HVAC Systems Noise Control Course No: M06-026

Credit: 6 PDH

A. Bhatia

Continuing Education and Development, Inc. 9 Greyridge Farm Court Stony Point, NY 10980 P: (877) 322-5800 F: (877) 322-4774 info@cedengineering.com

HVAC SYSTEMS NOISE CONTROL

Unwanted noise makes a workplace uncomfortable and less productive. When people

are surveyed about workplace comfort, their most prevalent complaints involve the

heating, ventilating and air-conditioning (HVAC) systems. The problems they cite most

frequently, aside from temperature control, have to do with excessive noise and

vibration.

There are practical and economical solutions to almost all noise problems in the built

environment. To approach the solution to any specific noise problem, we need to:

• Understand the basic principles of acoustics and how noise — unwanted sound

is produced, how it propagates, and how it is controlled.

• Learn the basics of noise control, and how to approach the problem from three

standpoints: the source of noise, the path it travels, and the point of reception.

• Become familiar with the treatments and modifications that can be applied to

HVAC system to reduce unwanted noise and vibration.

That’s what architects, engineers, contractors, and building owners — anyone

concerned with solving noise control problems in all types of buildings — will find in this

course. It includes information on how to solve specific noise control in HVAC systems.

The course is divided into 4 sections:

Section-1 The Fundamentals of Acoustics

Section-2 Noise Rating Methods

Section-3 Noise Descriptors

Section - 4 Controlling HVAC Noise and Vibration

Annexure -1 Rules of Thumb

Annexure -2 Glossary of Sound Terms

SECTION -1: THE FUNDAMENTALS OF ACOUSTICS

This section provides a brief introduction to the fundamentals of building acoustics that is

necessary to understand the material in the following sections. To emphasize the

simplicity of the approach, equations are kept to a minimum.

What is Sound and Noise?

Sound is a form of mechanical energy transmitted by vibration of the molecules of

whatever medium the sound is passing through. Noise may be defined as "unwanted or

undesired sound", which interferes with speech, concentration or sleep. To control noise

or sound, we need to know a little about its fundamental properties such as:

1. Frequency (pitch)

2. Wavelength

3. Amplitude (loudness)

Once these fundamental properties are understood, we can proceed to implement

effective noise control measures.

What is Sound Frequency?

The frequency or pitch of a sound wave is the number of times that its basic pattern

repeats itself per second. So, a musical note characterized by a pattern of pressure

variations that repeats itself 1200 times per second has a frequency of 1200 Hertz.

The frequency - cycles per second - of a sound is expressed in hertz - Hz. The

frequency can be expressed as:

f = 1 / T

Where

• f = frequency (s-1, Hz)

• T = time for completing one cycle (s)

Example

Calculate the time of one cycle for a 500 Hz tone.

T = 1 / (500 Hz)

= 0.002 s

Note - Sound and noise usually are not pure tones. Pure tone is described as a simple

vibration of single frequency.

The human ear can perceive sounds with frequencies ranging from 20 Hz up to 20000

Hz.

Wavelength

The wavelength of a sound wave is the distance between the start and end of a sound

wave cycle or the distance between two successive sound wave pressure peaks.

Numerically, it is equal to the speed of sound in the material such as air divided by the

frequency of the sound wave. For example:

The wavelength of a 100 Hz tone at room temperature is 1130* ft/sec divided by 100 Hz

which is equal to 11.3 ft.

*The speed of sound in air is approximately 1,130 feet per second.

Note - The higher the frequency the shorter shall be the wavelength.

λ = c / f

Where,

• λ = wavelength (m)

• c = speed of sound (m/s) (*in air at normal atmosphere and 0oC the sound of

speed is 331.2 m/s)

• f = frequency (s-1, Hz)

Example

Calculate the wavelength of a 500 Hz tone considering speed of sound as 331.2 m/s.

λ = (331.2 m/s) / (500 Hz)

= 0.662 m

Note - The velocity is the distance moved by the sound wave per second in a fixed

direction i.e. v = f x λ

Amplitude (loudness)

The amplitude of a sound wave is the height of the wave form. It is also the maximum

displacement for each air particle as it vibrates. The amplitude or loudness of a sound

wave is expressed by its sound pressure level. Sounds having the same wavelength

(equal frequency) may have differing loudness.

Note - Human ears distinguish one sound from another by its loudness and pitch.

Loudness is the amplitude or amount of sound energy (dB) reaching our ears. Pitch is

the speed of the vibrations or frequency used to identify the source of a sound. However,

both the loudness and pitch may vary depending upon where we are located relative to

the sound and the surrounding environment.

Sound Power and Sound Pressure

The difference between sound power and sound pressure is critical to the understanding

of acoustics.

Sound power is the amount of acoustical power a source radiates and is measured in

watts. Sound power is a fixed property of a machine irrespective of the distance and

environment. HVAC equipment generates sound power.

Sound pressure is related to how loud the sound is perceived to be to a human ear in a

particular environment. It depends upon the distance from the source as well as the

acoustical environment of the listener (room size, construction materials, reflecting

surfaces, etc.). Thus a particular noise source would be measured as producing different

sound pressures in different spots. Theoretically, the sound pressure “p” is the force of

sound on a surface area perpendicular to the direction of the sound. The SI-units for the

Sound Pressure are N/m2 or Pa.

The difference between the sound power that the unit generates and the sound pressure

that the ear hears is caused by the reduction (attenuation) of the sound along the path to

the room, and the reduction within the room (or "room effect") caused by the room

layout, construction and furnishings.

How Is Sound Measured?

The term used to describe sound is “decibel” (dB) – which is represented in Logarithmic

scale. Each 10 dB increase represents a tenfold increase in signal amplitude, while each

10 dB reduction represents a tenfold reduction.

What is logarithmic scale?

For instance, suppose we have two loudspeakers, the first playing a sound with power

W1, and another playing a louder version of the same sound with power W2, but

everything else (how far away, frequency) kept the same.

The difference in decibels between the two is defined to be:

• 10 log (W2/W1) dB where the log is to base 10

If the second produces twice as much power than the first, the difference in dB is:

• 10 log (W2/W1) = 10 log 2 = 3 dB

If the second had 10 times the power of the first, the difference in dB would be:

• 10 log (W2/W1) = 10 log 10 = 10 dB

If the second had a million times the power of the first, the difference in dB would be:

• 10 log (W2/W1) = 10 log 1000000 = 60 dB

This example shows one feature of decibel scales that is useful in discussing sound:

they can describe very big ratios using numbers of modest size. But note that the decibel

describes a ratio: so far we have not said what power either of the speakers radiates,

only the ratio of powers.

A decibel scale can be used for many purposes, but it is meaningful for a specific

purpose only when the reference level is defined. In acoustical work, two reference

levels are commonly used:1) Sound Power Level (Lw) and 2) Sound Pressure Level (Lp)

Sound Power Level (Lw) is a measure of sound energy output expressed in decibels

compared to a reference unit of 10-12 Watts. It is calculated by the formula:

Lw = 10 log10 (W source / W ref)

Where:

• Lw is sound power level in decibel (dB)

• W ref is 10-12 W;

• W source is sound power in W

If sound power increases by a factor of 2, this is equivalent to a 3 dB increase.

The use of sound power levels is the desirable method for equipment suppliers to furnish

information because sound pressure levels can then be computed in octave bands for

any desired reflective environment or space.

Sound Pressure Level (Lp)

The lowest sound pressure possible to hear (the threshold of hearing) is approximately 2

x 10-5 Pa (20 micro Pascal) and is given the value of 0 dB.

The threshold of pain in the ear corresponds to pressure fluctuations of about 200 Pa.

This second value is ten million times the first.

These unwieldy numbers are converted to more convenient ones using a logarithmic

scale (or the decibel scale) related to this lowest human hearable sound “pref” - 2 * 10-5

Pa, 0 dB.

Sound pressure level is the pressure of sound in relation to a fixed reference and is

described by a logarithmic comparison of sound pressure output by a source to a

reference sound source, P0 (2 x 10-5 Pa)

Lp = 10 log10 (P2 / P ref2)

Where:

• Lp = Sound Pressure Level (dB)

• Pref = 2 x 10-5 – Reference Sound Pressure (Pa);

• P = sound pressure (Pa)

Doubling the Sound Pressure raises the Sound Pressure Level with 6 dB (20 log (2)).

What does 0 dB mean?

This level occurs when the measured intensity is equal to the reference level. i.e., it is

the sound level corresponding to 0.02 mPa. In this case we have

Sound level = 20 log (Pmeasured/Preference) = 20 log 1 = 0 dB

So 0 dB does not mean no sound, it means a sound level where the sound pressure is

equal to that of the reference level. This is a small pressure, but not zero. It is also

possible to have negative sound levels: - 20 dB would mean a sound with pressure 10

times smaller than the reference pressure, i.e. 2 micro-Pascal.

Directionality of Sound Waves

Noise levels in rooms vary with distance from the source and with the properties of the

room.

In an outdoor situation, sound levels decrease as one moves away from the source.

Sound pressures decrease inversely proportional to the distance from the source. This is

true for a source that radiates sound approximately equally in all directions and where

there is only one path from the source to the receiver.

In a room, there are a very large number of possible paths from the source to the

receiver, involving various reflections off the room boundaries; the combination of all

these paths determines how sound behaves in the room. As a result, sound levels in a

room do not continue to decrease with increasing distance from the source for all

distances.

What measurements are useful?

Most of the sounds we hear are a combination of many different frequencies. Healthy

young human beings normally hear frequencies as low as about 20 Hz and as high as

20,000 Hz. The human ear however is most sensitive to sound in the frequency range

1000Hz to 4000 Hz than to sound at very low or high frequencies. This means that the

noise at high or low frequencies will not be as annoying as it would be when its energy is

concentrated in the middle frequencies. A higher sound pressure is therefore acceptable

at lower and higher frequencies. This knowledge is important in acoustic design and

sound measurement.

The engineer must clearly distinguish and understand the difference between sound

power level and sound pressure level. Even though both are expressed in dB, there is

no outright conversion between sound power level and sound pressure level. A constant

sound power output will result in significantly different sound pressures levels when the

source is placed in different environments.

It is much more useful to know the amount of sound power produced by a source than to

know the sound pressure measured under some particular condition; one should strive

to obtain sound power ratings of all potential noise sources at the design stage. The

acoustic engineer must take this into account when specifying noise levels.

Inverse Square Law

Most measuring instruments measure the average of the acoustic pressure over the time

period of the wave.Theoretically, the power in a sound wave goes as the square of the

pressure (similar to what electrical power goes as the square of the voltage). The log of

the square of x is just 2 log x, so this introduces a factor of 2 when we convert to

decibels for pressures. The difference in sound pressure level between two sounds with

p1 and p2 is therefore:

Lp = 10 log (p22/p1

2) dB or

Lp = 20 log (p2/p1) dB

The sound levels decrease 6 dB for each doubling of distance from the source. There

are however lots of things that have to be taken into account and understood before any

measurement would be meaningful.

Octave bands

Typical sources in buildings emit sound at many frequencies. Because sound occurs

over a range of frequencies, it is considerably more difficult to measure than temperature

or pressure. The sound must be measured at each frequency in order to understand

how it will be perceived in a particular environment. The human ear can perceive sounds

at frequencies ranging from 20 to 16,000 Hz, whereas, HVAC system designers

generally focus on sounds in the frequencies between 45 and 11,200 Hz. Despite this

reduced range, measuring a sound at each frequency would result in 11,156 data points.

For some types of analyses, it is advantageous to measure and display the sound at

each frequency over the entire range of frequencies being studied. This is called a full-

spectrum analysis.

To make the amount of data more manageable, this range of frequencies is typically

divided into smaller ranges called octave bands. Each octave band is defined such that

the highest frequency in the band is two times the lowest frequency. The octave band is

identified by its center frequency, which is calculated by taking the square root of the

product of the lowest and highest frequencies in the band.

The result is that this frequency range (45 to 11,200 Hz) is separated into eight octave

bands with center frequencies of 63, 125, 250, 500, 1,000, 2,000, 4,000, and 8,000 Hz.

For example, sounds that occur at the frequencies between 90 Hz and 180 Hz are

grouped together in the 125 Hz octave band.

Octave bands compress the range of frequencies between the upper and lower ends of

the band into a single value. A unit “Octave” defines the concept of these frequency

ranges and is the interval between two points where the frequency at the second point is

twice the frequency of the first.

Unfortunately, octave bands do not indicate that the human ear hears a difference

between an octave that contains a tone and one that does not, even when the overall

magnitude of both octaves is identical. Therefore, the process of logarithmically

summing sound measurements into octave bands, though practical, sacrifices valuable

information about the “character” of the sound.

Middle ground between octave-band analysis and full-spectrum analysis is provided by

one-third octave-band analysis. One-third octave bands divide the full octaves into

thirds. The upper cutoff frequency of each third octave is greater than the lower cutoff

frequency by a factor of the cube root of two (approximately 1.2599). If tones are

contained in the broadband sound, they will be more readily apparent in the third

octaves.

The use of octave bands is usually sufficient for rating the acoustical environment in a

given space. One-third octave bands are, however, more useful for product development

and troubleshooting acoustical problems.

Note - The measurement in building acoustics spectra are usually made in one-third

octave bands from about 100 Hz to 4000 Hz.

Sound Measurements

Sound is measured in decibels with the use of a Sound Level Meter. The sound level

meter responds to sound in approximately the same way as the human ear but it gives

an objective measurement of sound level. The sound to be measured is converted into

an electrical signal by a microphone. Normally condenser microphones are used for

precision grade instruments. Since the signal is quite small it is necessary to amplify it

and then convert for display on the meter.

When measuring the sound level from a source the meter should be held at a distance

of 1 meter from the source. The sound level in decibels A- scale can be measured at

each centre frequency using the filters and a graph can then be drawn for example as

shown below. This gives an indication of sound levels at different frequencies. This can

be useful information and can be used to design appropriate attenuation and acoustic

treatment at particular frequencies.

It will be seen from the above graph that the maximum noise output of 90 dB occurs at

4K (4000) Hertz. Also the lowest noise level of 50 dB occurs at a frequency of 63 Hz.

If sound attenuation is required for the machine in this example then the engineer would

concentrate at the high frequency noise spectrum (4K Hz). The insulation or proposed

attenuation system could be tailored especially for high frequency attenuation rather

than low frequency attenuation.

Is the Sound Pressures Levels additive from sounds emitting from various sources?

No

The decibel values of sounds can not be added or subtracted using the normal

arithmetic rules. Instead, decibel values must be converted back into absolute units of

power (Watts), when they can be added or subtracted directly before re-converting back

into decibels.

If N sources generating the same sound pressure level are combined, the overall sound

pressure level will be increased by 10 log N dB. Two sounds with average levels of 60

dB, for example, will together create a sound pressure level of 60 + 10 log 2 = 63 dB and

not 120 dB. The greater the difference in noise level between two noise sources, the

less effect there is on the combined level. Where the difference between two sources is

more than 6 dB, the combined level will be less than 1 dB higher than the louder source

alone. The table below shows the simple approximation method of obtaining the

equivalent combined noise level, when the dB difference between the two levels is

known.

Difference between two levels, dB

0

1 2 3 4 5 6 7 8 9 10 & more

Quantity to be added to the higher level

3 2.5 2 2 1.5 1.5 1 1 .5 .5 0

Examples of ways two sound levels contribute to each other based upon the difference between them

Procedure:

1) Select the highest level

2) Subtract the next highest level from the highest

3) Using the difference, go to the chart and find the addition to the highest level

4) Add this to the highest level

5) The result is the logarithmic sum of the two levels

Example # 1

If a fan has a sound level of 50dB and another, larger fan with a sound level of 55dB is

added, the difference between the two is 5dB, so 1dB is added to the higher figure,

giving a combined level of 56dB.

Example # 2

Combining Two Sound Spectrums

Band No. 1 2 3 4 5 6 7 8

Spectrum A 82 80 73 70 69 66 60 53

Spectrum B 79 77 71 68 67 64 59 52

Absolute Difference 3 3 2 2 2 2 1 1

Added to the highest 2 2 2 2 2 2 2.5 2.5

Example # 3

Combining an octave band spectrum into a single number

Band No. 1 2 3 4 5 6 7 8

Spectrum A 81 80 76 70 69 64 61 57

Procedure:

The highest number is 81, the next higher is 80; the difference is 1, the adder is 2.5, the

combined number is 83.5.

The next higher number is 76, the difference is 7.5, the adder is .5, and the combined

sum is 84

The next higher number is 70, the difference is 14, the adder is 0, and therefore the

combined number for this spectrum is 84 dB.

If two sound levels are identical, the combined sound is three dB higher than either. If

the difference is 10 dB, the highest sound level completely dominates and there is no

contribution by the lower sound level.

Equivalent Noise Level with “N” Sources Generating the Same Sound Pressure Level

The table below shows the simple approximation method of obtaining the equivalent

combined noise level, when the number of sources having same sound level are put in

the same room (i.e. difference between two sound levels is zero).

Number of

Sources

Increase in Sound Power

Level (dB)

Increase in Sound

Pressure Level (dB)

2 3 6

3 4.8 9.6

4 6 12

5 7 14

10 10 20

15 11.8 23.6

20 13 26

As an example, if one fan has a sound level of 50dB, and another similar fan is added,

the combined sound level of the two fans will be 53dB. If two more fans are added, i.e.

the sound source is doubled again; the resultant sound level from all four fans will be

56dB.

Average Ability to Perceive Changes in Noise Levels

Studies have shown that a 3 dB A increase is barely perceptible to the human ear,

whereas a change of 5 dB A is readily perceptible. The average ability of an individual to

perceive changes in noise levels is well documented (see table below). These guidelines

permit direct estimation of an individual’s probable perception of changes in noise levels.

As a general rule, an increase or decrease of 10 dBA in noise level is perceived by an

observer to be a doubling or halving of the sound, respectively.

Average Ability to Perceive Changes in Noise Levels

Change (dB A)

Human Perception of Sound

2-3 Barely perceptible

5 Readily noticeable

10 A double or halving of the loudness of sound

20 A dramatic change

40 Difference between a faintly audible sound and very loud sound

Sound Intensity

Sound Intensity is the Acoustic or Sound Power of a sound (W) per unit area in relation

to a fixed reference. The SI-units for Sound Intensity are W/m2.

The Sound Intensity Level can be expressed as:

LI = 10 log (I / Iref)

Where:

• LI = sound intensity level (dB)

• I = sound intensity (W/m2)

• Iref = 10-12 ----- reference sound intensity (W/m2)

The logarithmic sound intensity level scale matches the human sense of hearing.

Doubling the intensity increases the sound level with 3 dB (10 log (2)).

Example

The difference in intensity of 10-8 watts/m2 and 10-4 watts/m2 (10,000 units) can be

calculated in decibels as:

∆LI = 10 log ((10-4 watts/m2) / (10-12 watts/m2))

- 10 log ((10-8 watts/m2) / (10-12 watts/m2))

= 40 dB

Increasing the sound intensity by a factor of:

• 10 raises its level by 10 dB

• 100 raises its level by 20 dB

• 1,000 raises its level by 30 dB

• 10,000 raises its level by 40 dB and so on

Note! The sound intensity level may be difficult to measure, it is common to use “sound

pressure level” measured in decibels instead.

Sound Intensity and Sound Pressure

The connection between Sound Intensity and Sound Pressure can be expressed as:

I = p2 / ρ c

Where:

• p = sound pressure (Pa)

• ρ = 1.2 = density of air (kg/m3) at 20oC

• c = 340 – speed of sound (m/s)

Sound Power, Intensity and Distance to Source

The sound intensity decreases with distance to source. Intensity and distance can be

expressed as:

I = Lw / 4 π r2

Where:

• Lw = sound power (W)

• π = 3.14

• r = radius or distance from source (m)

To estimate the effect of distance from a sound source the rule of thumb calculation is

very simple. Sound normally weakens by 6dB each time the distance from the sound

source is doubled, i.e. a sound level measured as 60dB at 1 meter will be 54dB at 2

meters and 48 at 4 meters.

Smaller fan sound levels are sometimes measured at 1m, whereas the industry standard

is becoming standardized at 3m, for comparison purposes. The difference between 1m

and 3m is therefore 9dB. (41dB@3m) and 50dB@1m are in effect the same).

Note! The published noise levels are measured in dBA at a distance of 6 feet, the

industry standard.

Noise Control - Why do we need it?

1. Government Regulations enforced by Environment Protection Agency (EPA),

Occupation Safety and Health Administration (OSHA), Federal Housing

Administration (FHA), US Department of Housing and Urban development (HUD)

and other State and Local laws.

2. Insurance carriers apply pressure to prevent hearing loss claims filed under

workman's compensation laws.

3. Prolonged exposure to loud environment cuts down productivity; creates stress

and can lead to accidents.

4. Unwanted noise is nuisance. The most common problem in a room is too much

echo or reverberation. Too much echo can garble speech clarity and intelligibility.

______________________________________________________________________

SECTION -2 NOISE RATING METHODS

The human ear interprets sound in terms of loudness and pitch, while electronic sound

measuring equipment interprets sound in terms of pressure and frequency. This creates

a problem for measuring the impact of each sound on human hearing since each

frequency will be perceived differently by the ear. It is therefore important to equate and

adjust sound pressure and frequency to sound levels of human hearing. The goal has

been to develop a system of single-number descriptors to express both the intensity and

quality of a sound.

Although a number of procedures have been developed for rating the acceptability of

noise in buildings, three are most common; 1) A-weighted sound levels, 2) the noise

criteria (NC) and 3) the room criteria (RC) rating procedure.

A-weighted Measurement, dB (A)

The simplest method is A-weighted Measurement, dB (A) that combines octave-band

sound data into a single-number descriptor. As stated earlier, the sensitivity of the

human ear to sound depends on the frequency or pitch of the sound. People hear some

frequencies better than others. If a person hears two sounds of the same sound

pressure but different frequencies, one sound may appear louder than the other. This

occurs because people hear high frequency noise much better than low frequency noise.

Sound Level Meters are programmed to account for the human ear's sensitivity to

certain frequencies. The meter, measures the sound pressures in all the different

frequencies, corrects them for the human ears and then presents us with a single figure

representation of the noise – dB (A).

The letter “A” indicates that the sound has been filtered to reduce the strength of very

low and very high frequency sounds, much as the human ear does. The table below

summarizes common noise sources with their associated typical dB A.

A – Weighted Sound Level in Decibels (dB A)

Weighted Overall Level Noise Environment

120 Uncomfortably loud

(32 times as loud as 70 dBA)

Military Jet airplane takeoff at 50 feet

100 Very loud

(8 times as loud as 70 dBA)

• Jet flyover at 1000 feet

• Locomotive pass by at 100 feet

80 Very loud

(2 times as loud as 70 dBA)

• Propeller plane flyover at 1000 feet

• Diesel truck 40 mph at 50 feet

70 Moderately loud • Freeway at 50 feet from pavement edge morning hours

• Vacuum cleaner (indoor)

60 Relatively quiet

(1/2 as loud as 70 dBA) • Air-conditioning unit at 100 feet

• Dishwasher at 10 feet (indoor)

50 Quiet

(1/4 as loud as 70 dBA) • Large transformers

• Small private office (indoor)

40 Very quiet

(1/8 as loud as 70 dBA) • Birds calls

• Lowest limit of urban ambient sound

10 Extremely quiet

(1/64 as loud as 70 dBA)

Just audible

0 Threshold of hearing

The main advantages with the A -weighting are as follows:

1. It is adapted to the response of the human ear to sound

2. It is possible to measure easily with low cost instruments

“A-weighted” measurement is most commonly used for outdoor sound sources, such as

highway traffic or outdoor equipment. For an AHU, typically only the outside air, exhaust

air or casing radiated sound components would be rated in terms of dBA, not the ducted

sound components. Sound power is not typically rated in dBA because the A-weighting

network is intended to rate the annoyance or “effect” of outdoor noise sources, and it is a

characteristic of the noise source, not the effect of the noise.

Limitations

Speech happens predominantly in the mid-frequencies. But if you take readings in dBA

there may be a huge low frequency component happening that you don’t know about,

because dBA is a weighted average. Therefore measuring average spectra, such as

dBA, typically doesn’t reveal the complete spectral balance of sound and do not

correlate well with the annoyance caused by noises. Different sounds can receive the

same rating but retain dissimilar subjective qualities. A-weighted sounds are best for a

comparison between sounds that are similar but differ in level, such as comparing the

loudness between two different makes of a fan. Therefore, the A-weighted sound level is

not the best tool for measuring HVAC systems as a whole, but it is better used for

measuring a single component like a fan.

dB (A) and dB (C)

Similar to A-weighted measurement is C-weighted measurement denoted as dB (C). The

letters “A” or “C” following the abbreviation “dB” designate a frequency-response function

that filters the sounds that are picked up by the microphone in the sound level meter.

• The A weighting filters out the low frequencies and slightly emphasizes the upper

middle frequencies around 2-3 kHz. A-weighting, which is most appropriately

used for low-volume (or quiet) sound levels, best approximates human response

to sound in the range where no hearing protection is needed.

• By comparison C weighting is almost unweighted, or no filtering at all. C-

weighting is used for high volume (or loud) sound levels where the response of

the ear is relatively flat.

A-weighting is used to measure hearing risk and for compliance with OSHA regulations

that specify permissible noise exposures in terms of a time-weighted average sound

level or daily noise dose. C-weighting is used in conjunction with A - weighting for certain

computations involving computation of hearing protector attenuation.

Noise Criterion – NC values

In acoustic analysis, the air conditioning designer is concerned with what the human ear

perceives. The "loudness" of the sound is based not so much on absolute numbers, as

on the human ear's perception. The human ear responds differently to loudness at

different frequencies.

Noise criteria (NC) curves plotted on a scale of Frequencies (Octaves) vs. Decibels

(Loudness in dB); represent approximately equal loudness levels to the human ear.

These curves define the limits that the octave band spectrum must not exceed. For

instance, to achieve NC-35 rating, the sound spectrum must be lower than the curve in

every octave band.

NC is particularly used in Heating, Ventilating and Air Conditioning (HVAC) work, with an

NC-35 being the most common requirement.

NC Level – Methodology

The following steps describe how to calculate an NC rating:

1. Plot the octave-band sound-pressure levels on the NC chart.

2. The highest curve crossed by the plotted data determines the NC rating.

The sound is measured in octave bands from 63 to 8000 Hz and plotted against a set of

curves. The point where the spectrum touches the highest point tangent to NC curve,

determines the NC rating for the spectrum. The curves for a given rating allow less

sound with increasing frequency. The rating numbers correspond to the curve level in

the 1000-2000 Hz octaves. Two spectra can have the same NC value but quite different

shapes. Although the relationship between the A-weighted level and the NC number will

depend on the actual spectrum, the requirements are that the sound pressures

measured at each octave band must be below the specified NC curve (within a 2dB

tolerance) if they are to meet the NC rating.

Example

Calculate the NC level and the sound insulation needed for an office with a nearby noise

source as follows:

Noise Source Spectrum

Octave Band Centre Frequency (Hz) 63 125 250 500 1K 2K 4K 8K

Measured Sound Levels (dB) 69 63 64 62 58 57 55 51

The noise source is plotted on the NC curves below.

The closest fit NC curve to the noise source is NC 58.

The generally accepted sound level for office spaces is NC35 to NC 45, therefore if say

NC 40 is chosen, then the amount of insulation at each frequency can be calculated.

Noise Source Spectrum

Octave Band Centre Frequency (Hz) 63 125 250 500 1K 2K 4K 8K

Measured Sound Levels (dB) 69 63 64 62 58 57 55 51

NC 40 from Figure above (dB) 63 56 50 45 42 40 38 37

Insulation required (dB) 69 – 63 = 6 7 14 17 16 17 17 14

Limitations

While NC is a great improvement over dB (A) ratings, one needs to keep its limitations in

mind.

It gives little indication of the sound character (quality). Two different sounding noise

spectra with different acceptance from people may be rated at the same NC level. For

example, equipment with a dominant single low frequency peak will sound much more

offensive than equipment with a spectrum that more closely matches the NC curve. For

HVAC equipment especially package and self contained units, it is important to compare

the noise generated in the first (63 Hz) and second (125 Hz) octave bands. Higher noise

in these octave bands can cause a rumble in the conditioned space.

NC curves do not account for sound at very low frequencies - the 16 Hz and 31.5 Hz

octave bands. Although HVAC equipment manufacturers typically do not provide data in

these bands (because it is very difficult to obtain reliably), these octave bands do affect

the acoustical comfort of the occupied space. Higher frequency sound, which is much

easier to attenuate with acoustic insulation is reduced significantly and is unable to mask

unwanted sound. Low frequency noise is attenuated much less, causing the annoying

rumble.

NC curves are not ideal background spectra to be sought after in rooms to guarantee

occupant satisfaction but are primarily a method of rating the noise level. There is no

generally accepted method of rating the subjective acceptability of the spectral and time-

varying characteristics of ambient sounds. In fact, ambient noise having exactly an NC

spectrum is likely to be described as both rumbly and hissy, and will probably cause

some annoyance. NC ratings are often used to establish maximum noise levels

associated with HVAC machinery.

Relationship between A-weighted and NC

Although the relationship between the A-weighted level and the NC number will depend

on the actual spectrum, usually the A-weighted level is about 7 dB greater than the NC

value.

Room Criteria (RC)

The room criterion (RC) method is currently the most favored method for determining

sound levels of HVAC systems noise. The RC system overcomes the shortcomings of

the NC system to some extent. The major difference is that RC curves give an additional

indication of sound character. The RC system also uses the 16.5 and 31 Hz bands,

which allows the criteria to account for acoustically produced vibration in light building

construction. However, the RC calculations should only be performed when valid data

exists. If the 16.5 and 31 HZ data are not available, it should not be extrapolated from

available values for RC calculations.

In addition, the RC rating consists of two descriptors. The first descriptor is a number

representing the speech interference level (SIL) of the sound. The second descriptor is a

letter denoting the character of the sound as a subjective observer might describe it.

• The letter N identifies a neutral or balanced spectrum - this sound will not have

any identity with frequency and will sound bland.

• The letter R denotes “rumble” - this occurs whenever any specific RC value is

more than 5 dB greater than the standard curve values in the 500 Hz and below

octave bands.

• The letter H indicates “hiss” - this occurs whenever any specific RC value is more

than 3 dB greater than the standard curve values above the 500 Hz octave band.

• The letter V represents “vibration”

Note the Facts below:

a. RC measures background noise in the building over the frequency range of 16-

4000 Hz. This rating system requires determination of the mid-frequency average

level and determining the perceived balance between high-frequency (HF) sound

and low-frequency (LF) sound. Studies show that an RC between 35 and 45 will

usually provide speech privacy in open-plan offices, while a value below 35 does

not. Above RC–45, the sound is likely to interfere with speech communication.

b. The arithmetic average of the sound levels in the 500-, 1,000- and 2,000-Hz

octave bands is 44.6 dB; therefore, the RC 45 curve is selected as the reference

for spectrum quality evaluation.

c. The sound is rated rumbly, if any of the bands from 31.5 Hz through 250 Hz are 5

dB or more higher than the RC curve.

d. The sound is rated as a hiss, if the 2,000- of 4,000-Hz bands are greater than 3

dB above the RC curve.

RC method is a preferred rating system advocated by American Society of Heating,

Refrigerating and Air-Conditioning Engineers and can be further looked at in the

ASHRAE application handbook.

Summarizing….

When selecting a rating method it is important to consider how the rating will be used.

All three dB (A), NC and RC methods consider speech interference but A-weighted

method is primarily used for outdoor noise environment.

While NC doesn’t provide any quality assessment and don’t evaluate low frequency

rumble, RC method evaluates sound quality and provides some diagnostic capability.

NC method is used to rate components such as air terminal diffusers; RC method is not

recommended to be used to rate components.

The RC method is considered to be the better measure of sound between the two

methods and is slowly replacing the NC method as a means of analyzing sound.

______________________________________________________________________

SECTION -3 NOISE DESCRIPTORS

Acoustical design for buildings requires quantitative information about acoustical

products, materials and systems so that recommended design criteria can be met. For

most noise control work in buildings, the two most important acoustical properties of the

materials and systems used are “sound-absorption” and “reverberation”. The term sound

absorption is most often used to describe single space acoustic applications where

reverberation of noise or echo is the main problem. Projects such as indoor arenas,

sporting facilities, cinemas, theatres and studios need to be designed to minimize this

type of reverberant noise.

This section provides information on the common noise descriptors.

Reverberation

The reverberation time (RT), widely used as a design parameter in architectural

acoustics, is the time required for a loud sound to fade away to inaudibility after the

source has been turned off. More precisely, it indicates how long it takes until the sound

pressure level in a room is decreased by 60 dB after the sound source is terminated.

Excessively reverberant rooms are usually noisy and speech communication may be

interfered with. The reverberant sound level and the RT in the room can be reduced

through the installation of appropriate sound absorbing materials.

Factors Influencing Reverberation Time

Reverberation time is affected by the size of the space and the amount of reflective or

absorptive surfaces within the space.

• Size of space - Reverberation time is directly related to the room volume. In

general, larger spaces have longer reverberation times than smaller spaces.

Therefore, a large space will require more absorption to achieve the same

reverberation time as a smaller space.

• Amount of absorptive surfaces -Highly absorbent surfaces shorten the

reverberation time whereas highly reflective surfaces lengthen the reverberation

time.

Figure 1 (left) below gives optimum reverberation time for speech versus room volume.

As the room volume increases, the optimum reverberation time also increases. Since

this contour indicates a unique optimum reverberation time for each room volume, it is

possible to calculate the related optimum total absorption necessary for each room

volume.

Figure 2 (right) above shows the resulting range of acceptable total room absorption (in

metric sabins) versus room volume (in cubic metros) that would lead to the desired

optimum reverberation time within +0.1 second. Designing for a total room absorption as

close as possible to the middle of this range should lead to nearly optimum conditions for

speech.

How to calculate Reverberation Time?

There are several formulas for calculating reverberation time; the most common formula

is the Sabin Formula. The formula is based on the volume of the space and the total

amount of absorption within a space.

The Reverberation Time - Ta - can be expressed as:

Ta = 0.16 V / A (1)

Where

• 0.16 is an empirical constant

• Ta = reverberation time (s)

• V = room volume (m3)

A = Total sound absorption of the room (room surface area x average absorption

coefficient + absorption of furniture/people, m2).

The absorption of a room is obtained by summing the absorption of all the surfaces in

the room, i.e. walls, ceiling, floor and all the furniture in the room. The total amount of

absorption within a space is referred to as sabins.

What is Sabin?

The sound absorption for a sample of material or an object is measured in sabins or

metric sabins. One sabin maybe thought of as the absorption of unit area (1 m2 or 1 ft2)

of a surface that has an absorption coefficient of 1.0 (100%). When areas are measured

in square meters, the term metric sabin is used. The absorption for a surface can be

found by multiplying its area by its absorption coefficient. Thus for a material with an

absorption coefficient of 0.5, 10 ft2 of this material has a sound absorption of 5 sabins

and 100 m2, of 50 metric sabins.

BASIC CLASSES OF MATERIALS

Two classes of materials are often used for engineering noise control:

1. Sound absorption materials are the materials that absorb sound; reduce

reflections from surfaces and decrease reverberation within spaces. Sound

absorbing materials are usually fibrous, lightweight and porous.

2. Sound partitions or barriers reduce sound transmission between adjacent

spaces. These materials are usually non-porous and a good reflector of sound.

A sound barrier material is a poor absorber of sound and an absorbent material is a poor

barrier.

Sound Absorbing Materials

The most common types of absorbing materials are rock wool, fiberglass, cloth,

polyurethane and cellulose fibers. All these materials are different in a sense that these

do not absorb sound of all frequencies equally well. Sound absorption is influenced by

factors such as material density thickness and, in the case of fibrous insulation products,

fiber size and diameter, generally fine small diameter fibers will give superior absorption

than coarser fiber blends.

These are characterized by high absorption coefficients at high frequencies, decreasing

at lower frequencies depending on the type and thickness of the material.

Sound Absorbing Coefficient (SAC)

The sound absorption coefficient indicates how much of the sound is absorbed by the

material over a range of frequencies.

Example: ½” drywall on 2x4 studs has an absorption coefficient at 125 Hz of 0.29.

The figure below illustrates typical absorption coefficient values versus frequency for 2.5

and 5.0 cm thick porous samples. Such absorbing materials function by resisting the air

flow associated with the acoustical vibrations of the air and are most effective at higher

frequencies, where they are thicker relative to the wavelength of the sound.

Since absorption decreases at lower frequencies; porous absorbing materials a few

centimeters thick will never be highly absorptive at lower frequencies. Increased low

frequency absorption can be obtained by adding an air space between the material and

the rigid backing so that the sound absorber behaves like a thicker material.

There are three main standard methods used to test materials for absorption. Two of

them are reverberation chamber methods – ASTM C423 in the U.S.A. and ISO 354 in

Europe. These two methods are quite similar, but the ISO method – in general – will

produce slightly lower overall numbers than the ASTM method. The other method is the

impedance tube method, or ASTM C384. This method places a small sample of the

material under test at the end of a tube and measures the absorption. Again, the

numbers from this test are usually lower and are also not as representative of real-world

applications of materials relative to the reverberation chamber methods.

Room Absorption Characteristics

The total sound absorption in a room can be expressed as:

A = S1 α1 + S2 α2 +... + Sn αn = Σ Si αi

Where:

• A = the absorption of the room (m2 sabin)

• Sn = area of the actual surface (m2)

• αn = absorption coefficient of the actual surface

It is important to note that the absorption and surface area must be considered for every

material within a space in order to calculate sabins.

The sound absorption coefficient can be expressed as the ratio of the absorbed sound

energy to the incident energy or can also be expressed as a percentage:

α = Ia / Ii

Where:

• Ia = absorbed sound intensity (W/m2)

• Ii = incident sound intensity (W/m2)

The absorption coefficient varies with frequency of sound and the angle of incidence of

the sound waves on the material. These coefficients are normally determined for one-

third octave bands under standard conditions. This provides an average value for each

one-third-octave band and all angles of incidence. Data are usually given at six standard

frequencies (125, 250, 500, 1000, 2000 and 4000 Hz), even more modern laboratories

will provide data at all one-third-octave bands from about 100 Hz to 5000 Hz.

Absorption coefficients are usually obtained from manufacturers' literature and are

usually obtained by testing in a reverberation test room (a room which has very long

Reverberation Time) and then measure the RT so the coefficient can be derived from

Sabin equation (the original version of RT calculation). There is a standard that details

this procedure. The value of the coefficient for the same material varies with the type of

the mounting in the test room.

Mean Absorption Coefficient

The mean absorption coefficient for the room can be expressed as:

am = A / S

Where:

• am = mean absorption coefficient

• A = the absorption of the room (m2 sabin)

• S = total surface in the room (m2)

A rooms acoustic characteristics can be calculated with the formulas above, or

estimated for typical rooms.

The table below gives mean sound absorption coefficient values and reverberation time

for some typical rooms.

Typical Room Room Characteristics

Reverberation Time

Mean Sound Absorption Coefficient

Radio and TV

studio Very Soft 0.40 0.2 < Ta < 0.25

Restaurant

Theater

Lecture hall

Soft 0.25 0.4 < Ta < 0.5

Office, Library,

Flat Normal 0.15 0.9 < Ta < 1.1

Hospital,

Church Hard 0.10 1.8 < Ta < 2.2

Large church,

Factory Very Hard 0.05 2.5 < Ta < 4.5

The mean sound absorption coefficient should be calculated when more accurate values

are needed.

Noise Reduction Coefficient (NRC)

In sound absorption applications the term NRC is often used as a performance indicator

of how absorptive a measure is. The higher the NRC, the greater will be the sound

absorption at those frequencies.

Example: ½” gypsum board (“drywall”) on 2x4 studs has an NRC of 0.05. Soft materials

like acoustic foam, fiberglass, fabric, carpeting, etc. will have high NRCs; harder

materials like brick, tile and drywall will have lower NRCs.

A material’s NRC is an arithmetic average of the four sound-absorption coefficients at

250, 500, 1000 and 2000 Hz rounded to the nearest 5%. A material with NRC of 0.80 will

absorb 80% of the sound that comes into contact with it and reflect 20% of the sound

back into the space. NRC is useful for a general comparison of materials. However, for

materials with very similar NRCs, it is more important to compare absorption coefficients.

NRC Limitations – What you should know

1. The noise reduction coefficient (NRC) is only the average of the mid-frequency

sound absorption coefficients (250, 500, 1000 and 2000 Hz).

2. NRC does not rate sound absorption at low frequencies less than 250hz, which is

often the most critical to the application.

3. NRC does not address material’s barrier effect.

4. NRC single number rating system is convenient for ranking, the average

effectiveness of different materials. However, for a more complete acoustical design,

it is necessary to consider individual coefficients at each frequency.

PARTITIONS, BARRIERS & ENCLOSURES

Partitions, barriers and enclosures are the products used for restricting the passage of

an airborne acoustic disturbance from one side to the other. These are nonporous and

dense materials often having the structural properties. Typical barrier materials are steel,

lead, plywood, gypsum wallboard, glass, brick, concrete, concrete bloc etc.

The quantity used to rate a sound attenuating material is transmission loss in dB. The

sound transmission loss for different materials is given in the table below:

Sound Transmission Loss (dB) Material Thickness (mm)

Surface Density (kg/m3) 125 Hz 500 Hz 2000 Hz

Plaster Brick Wall

125 240 36 40 54

Compressed Strawboard

56 25 22 27 35

Acoustic panel (Sandwiched type steel sheet with fiber glass)

50 27 19 31 44

Chipboard 19 11 17 25 26

Plaster board

9 7 15 24 32

Plywood 6 3.5 9 16 27

Sound Transmission Loss (STL)

When sound impinges on a partition, part of it is transmitted, some is reflected and the

rest is absorbed. The difference between the sound power incident on one side of the

partition and that radiated from the other side (both expressed in decibels) is called the

sound transmission loss (TL) in dB. The larger the sound transmission loss (in decibels),

the smaller will be the amount of sound energy passing through the partition.

The Mass Law is the general rule for determining the transmission loss of a partition.

TL = 20 log (fm) – 47.5

Where:

• f is the frequency of sound wave, Hz

• m is the surface density of material, kg/m2

Based on this equation, the sound insulation will increase by 6 dB for either a doubling of

the frequency or surface density. In practice, however, the actual increase in

transmission loss is somewhat less than that predicted by the Mass Law.

Sound Transmission Class (STC)

The Sound Transmission Class (STC) STC is a single-number rating of how effective a

material or partition is at isolating sound.

Example: ½” drywall has an STC of 28. The larger the STC value, the better the

partition, i.e., the less sound energy passes through it. For example, loud speech can be

understood fairly well through an STC 30 wall but should not be audible through an STC

60 wall.

Hard materials like rubberized sound barriers, concrete, brick and drywall will have high

STCs. Softer materials like mineral fiber, acoustic foam and carpet will have much lower

STCs. Virtually every material filters out some of the sound that travels through it, but

dense materials are much better at this than are porous or fibrous materials. Like NRC,

STC is useful to get an overview-type comparison of one material or partition to another.

However, to truly compare performance, the transmission loss numbers should be

reviewed.

The transmission of sound between rooms involves not only the direct path through the

separating assembly, but the flanking paths around the assembly as well. Flanking paths

are the means for sound to transfer from one space to another other than through the

wall. Sound can flank over, under, or around a wall. Sound can also travel through

common ductwork, plumbing or corridors. Special consideration must be given to spaces

where the noise transfer concern is other than speech, such as mechanical equipment

or music.

Recommended Ratings

In general, loud speech can be understood fairly well through an STC 30 wall but should

not be audible through an STC 60 wall. An STC of 50 is a common building standard

and blocks approximately 50 dB from transmitting through the partition. Constructions

with a higher STC (as much as 10dB better - STC 60) should be specified in sensitive

areas where sound transmission is a concern.

The Uniform Building Code (UBC) contains requirements for sound isolation for dwelling

units in Group-R occupancies (including hotels, motels, apartments, condominiums,

monasteries and convents). UBC requirements for walls, floor/ceiling assemblies: STC

rating of 50 (if tested in a laboratory) or 45 (if tested in the field).

STC Limitations – What you should know

1) The STC rating is based on performance with frequencies from 125 to 4000 Hertz

(the speech frequencies) and does not access the low frequency sound transfer. The

rating provides no evaluation of the barrier's ability to block low frequency noise,

such as the bass in music or the noise of some mechanical equipment.

2) The STC rating is a lab test that does not take into consideration weak points,

penetrations, or flanking paths. The field test however, does evaluate the entire

assembly and includes all sound paths.

3) Improving the STC rating of a wall will probably not affect the reverberation or

reflections within the space.

NRC V/s STC

NRC and STC are completely exclusive of each other.

Acoustic wall treatment with a high NRC can stop sound reflecting back into the space,

possibly lowering the noise level within the space whereas acoustic wall treatments will

not stop sound from passing through and into an adjacent space; therefore they do not

improve the STC.

Improving the STC rating of a wall (adding mass, air space, cavity, insulation etc) will

reduce the noise transfer to the adjacent space whereas improving the STC rating of a

wall will probably not affect the reverberation or reflections within the space.

It is important to understand the difference between sound-absorption and sound

transmission loss. Materials that prevent the passage of sound are usually solid, fairly

heavy and non-porous. A good sound absorber is 15 mm of glass fiber; a good sound

barrier is 150 mm of poured concrete.

______________________________________________________________________

SECTION -4 HVAC NOISE CONTROL

In the majority of rooms or open spaces, the most common source of disturbing noise is

the HVAC noise. The primary objective for acoustical design of HVAC systems and

equipment is to ensure that the acoustical environment in a given space is not

unacceptably affected by HVAC system-related noise or vibration.

The ASHRAE Handbook, Chapter 43: “Sound and Vibration Control provides guidelines

and procedures for noise control in HVAC systems. We won’t repeat the detailed

procedures here, but present the basic principles of the design process for the building

designers to grasp the general concepts. This section will discuss 25 key noise reduction

strategies in HVAC Systems.

Noise Points in HVAC Systems

Sound and vibration are created by a source, are transmitted along one or more paths

and reach a receiver. It is important to know which of these paths is most likely to cause

acoustical problems so that it can be dealt with in the design phase of the project.

The design of ducted HVAC systems must address five distinct but related issues.

Treatments and modifications can be applied to any or all of these elements to reduce

unwanted noise and vibration, although it is usually most effective and least expensive to

reduce noise at the source.

Duct borne Noise

Duct borne noise is generated by the flow of air and is directly dependent on the velocity

of air. This component of the noise (sometimes called regenerated noise) propagates

along the ductwork, follows all transitions and takeoffs, and ultimately exits at the diffuser

or grille. The noise increases very rapidly as the flow velocity is increased and/or by

changes in cross-section in the duct. It is affected by:

1. Objects such as dampers, grilles, registers, etc;

2. Constrictions in duct cross sectional area, silencer splitters etc;

3. Jet noise, inlet or discharge noise through orifices and air flow around bends and

duct take offs (branches).

To avoid duct borne noise, note the following:

1. Select quiet fans based on sound power data and use variable frequency drives

preferably. Provide good fan outlet conditions in accordance with ASHRAE

guidelines;

2. Route duct in accordance with SMACNA guidelines. Do not turn the air in “the

wrong direction.

Radiated Equipment Noise

Radiated equipment noise is generated by vibration of the mechanical equipment such

as pump, compressor, air handler, fan and motor and is transmitted through the wall or

floor into the adjacent space. Because the natural frequencies of floors, walls, beams,

and columns typically range from 10 to 60 Hz, these components can be excited into

greater vibration magnitude when located near equipment operating in a matching

frequency range. The noise may be transmitted directly or via the duct to the building

structure, or it may be radiated directly from the duct.

Excessive vibration and noise may occur due to:

1. Damaged or unbalanced fan wheel;

2. Belts too loose; worn or oily belts;

3. Speed too high;

4. Incorrect direction of rotation;

5. Improper lubrication.

To reduce vibration noise paths, note the following:

1. The fan and duct should be isolated from the building structure;

2. Use flexible (isolating) coupling between the fan and duct;

3. Use of vibration isolation foot pads below the rotating equipment;

4. Use of inertial base resilient mount floating floor;

5. Sound absorption treatment of the mechanical room walls and roof;

6. Routine maintenance of fan drive and motor.

Duct Break- In and Break - Out Noise

Duct breakout noise transmits through the wall of the duct, thus impacting the adjacent

space. Noise breakout from ducts can occur from:

1. Fan noise passing through the duct;

2. Aerodynamic noise (also known as regenerated noise), from obstructions fittings

etc in the duct;

3. Turbulent airflow causing duct walls to vibrate and rumble radiating low

frequency airborne noise

To reduce break-out noise, note the following:

1. Design for low velocity. The air velocity in the main duct should not exceed 1500

fpm and velocity through the branch ducts should be limited to 600 fpm;

2. Design proper duct fittings for smooth flow and gradual velocity changes;

3. Make ducts stiffer. Use heavier material for duct walls and increasing damping

(i.e. thicker steel sheeting). External bracing of ducts can also increase stiffness;

4. Adding damping (spray on or self adhesive compounds);

5. Add insulation on the external duct surfaces;

6. Provide acoustic silencers to reduce fan noise prior to allowing a duct to run

above the acoustical tile ceiling of an occupied space.

Duct break-in noise is radiated equipment noise that enters the ductwork and

propagates down the duct system. Flexible ducts are especially prone to break-in noise

because of light weight.

To minimize break-in noise, note the following:

1. Avoid flexible ducts. Replace lightweight flexible ducts with heavier ducting such

as sheet steel.

2. The flexible ducts can be enclosed in a solid enclosure constructed from timber,

plasterboard or sheet steel, etc.

3. Add acoustic lining either glasswool or rockwool at least 1 inch thick;

4. Before enclosing flexible ducts, it should be noted that noise in the ceiling cavity

will most likely penetrate the ceiling. This will happen more so if lightweight lay-in

tiles are used. Fixed plasterboard ceilings give better acoustic performance than

lightweight ceiling tiles.

Terminal Noise

The final links in the air distribution chain are the terminal air devices. These are the

“grilles,” “diffusers”, “registers” and “vent covers” that go over the duct opening in the

room. When selecting terminal devices; always select a device that has “noise criteria”

rating of NC-30 or lower for the designed airflow rate.

NOISE REDUCTION STRATEGIES IN HVAC SYSTEMS

In HVAC systems, the source of noise is a combination of different processes, such as

mechanical noise from fan(s), pump(s), compressor(s), motor(s), control dampers, VAV

boxes and air outlets such as diffusers, grilles, dampers and registers. The HVAC noise

that ultimately reach indoor space is made up of:

a. Low-frequency fan noise - Centrifugal fans generate highest noise in the low

frequency range in or below the octave centered at 250-Hz.

b. Mid-frequency airflow or turbulence-generated noise – Terminal units, variable-

air-volume (VAV) boxes produce their highest noise levels in the mid frequency

range in the octaves centered at 250, 500 and 1000 Hz bands.

c. High-frequency damper and diffuser noises – Diffusers, and grilles, however,

typically generate the highest noise levels in the octaves centered at 1000 Hz, or

above.

It is important to appreciate that the sound energy can travel by several different paths.

The airborne sound energy from the fan can propagate through the duct system in both

directions from the fan, as well as into the fan room from the fan casing. Needless to

say, each of these issues must be addressed, or else the design will fail.

The vibration energy of the fan can propagate through the fan room floor and other parts

of the building structure, as well as through the walls of the duct system.

Note: High frequency noise can be reduced using passive devices (attenuators, lining

etc), but noise components at frequencies below 400-500 Hz are most difficult to

attenuate.

We will discuss various noise attenuation techniques in following paragraphs.

LOCATION OF EQUIPMENT

The installation position of equipment (chillers, air cooled condensers, cooling towers,

exhaust fans etc.) is of critical importance in determining the noise level at the affected

noise sensitive receivers. Where practicable, the equipment should be placed in a plant

room with thick walls or at a much greater distance from the receiver or behind some

large enough obstruction (e.g. a building or a barrier) such that the line of sight between

receiver and the equipment is blocked. If noisy equipment has to be placed near a

receiver due to spatial or other constraints, consider acoustic enclosures and barriers.

Air handlers are typically housed in mechanical rooms within the indoor space. These

mechanical equipment rooms (MER) should be located away from sensitive areas and

never on a roof directly over a critical space. If possible, isolate the equipment room by

locating elevator cores, stairwells, rest rooms, storage rooms and corridors around its

perimeter. The walls, floors and doors of MER must have high sound reduction indices

and as the airborne sound easily passes through small gaps and cracks, the penetration

points for pipes, cables and ducts through the walls must be well sealed. The doors must

also be fitted with rubber sealing strips.

MER Size

As a rule, the larger the MER room, the quieter the HVAC system will be. It is important

to have a sufficiently spacious mechanical room so that ductwork can be routed to

accommodate insulation material and silencers. The air handler units and other

mechanical equipment should be placed away from the walls or ceilings. There is a

phenomenon called “close coupling,” in which a small air space will conduct cabinet

vibratory motion to the wall or ceiling. A space of approximately 3 feet usually suffices.

Provide a nominal 4 inch concrete housekeeping pad beneath equipment cabinets to

minimize the effects of close coupling to the floor.

MER Wall Construction

The desired noise levels in the adjacent space and the amount of airflow determine the

requirements for MER wall construction/treatment.

One method is to acoustically insulate the mechanical room walls and ceilings with fiber

insulation mounted on wooden battens. The insulation shall be clad with perforated

Aluminum sheet of 36 gauge. When sound strikes a surface, some of it is absorbed,

some of it is reflected and some of it is transmitted through the surface. Dense surfaces,

for the most part, will isolate sound well, but reflect sound back into the room. Porous

surfaces, for the most part, will absorb sound well, but will not isolate. Acoustic insulation

reduces the level of noise because while the sound waves are reflected back and forth in

the room and interact with the sound-absorbing materials to lose some energy each

time.

Wherever possible, plan the entrance to the MER from a non-critical buffer space such

as a freight elevator lobby or janitor’s closet. If this is done, a solid core wood door or

hollow metal door with glass fiber packing may be used. The door should have quality

gaskets and drop seals to form an airtight seal all around.

MER Equipment

Right sizing of HVAC equipment and distribution system matching the turndown

capabilities to the load being served is in essence, a fundamental engineering

component to sustainable design. Besides high possibility of higher noise levels at

source, an oversized equipment will short-cycle, frequently switching on & off which

mean startup noise every time the equipment kicks in. Choosing the correct size

(heating and/or cooling output) is critical not only in getting the best acoustic efficiency

but also comfort, and lowest maintenance and operating costs over the life of the new

system.

Further, the acoustical capabilities of the mechanical equipment should match the

acoustical capabilities of the MER wall construction. For example, if a standard

centrifugal fan system is selected for the project, and the acoustical engineer projects

noise levels approximating NC 45 in the occupied space directly adjacent, the

requirements for the wall and door should be in line with the known sound spectrum

within the equipment room and projected noise levels outside the equipment room. A

different MER wall may be selected for a mixed flow air handler and NC 35 projections.

On average, quieter equipment may generally be more expensive. However, it is almost

always more economical in the long run to buy quieter equipment than to reduce noise

by modification after purchase. Most equipment has a range of readily available noise

control devices that are able to deal with the noise problems. It is advisable that noise

levels specification is included when ordering new equipment. This allows the equipment

suppliers to select appropriate equipment and optional noise control devices to suit the

acoustic requirements.

MER Duct penetrations

All penetrations of MER walls must be sealed airtight. Sound, like air and water, will get

through any small gap. All wall constructions must extend and sealed up to the floor

construction above. If the wall is located under a beam, the space between the top of the

beam and the deck must also be sealed.

The supply duct penetration must also remain resilient, avoiding rigid contact between

the duct and the wall. The space between the duct and the wall should be packed with

glass fiber insulation and be closed to a 1/2” wide gap. This gap should then be sealed

with a permanently resilient sealant such as silicone caulk or acoustical sealant.

Duct penetration detail at wall

The figure above illustrates a recommended means of sealing the duct at the MER wall.

The key to success is to allow no direct contact of the duct to the equipment room wall

and to leave no voids between the ductwork and the wall.

VIBRATION ISOLATION

Rotating or motor-driven machinery generates lot of vibration that can travel through a

building's structure and radiate from the walls, floors and ceilings in the form of airborne

noise. It is therefore necessary to isolate any vibrating equipment from the surrounding

structure. This can be accomplished by:

1. Balancing the machines to reduce the forces at the source; a well-balanced

motor and fan impeller at as low a RPM as possible shall result in lower

vibrations;

2. Using resilient, anti-vibration mountings between the machine and the supporting

structure. Common materials used as vibration isolators are rubber, cork, various

types of steel springs, and glass fiber pads;

3. Placing the rotating equipment on grade, in the basement, or in the sub-

basement. Most pumps need isolators even in these desirable locations;

4. Using inertia blocks at the base of the equipment to reduce motion, lower the

center of gravity, minimize the effect of unequal weight distribution of the support

equipment, and stabilize the entire vibration isolation system. Generally an inertia

block should be at least 6” thick and very stiff and rigid to avoid significant flexure

in any direction;

5. Using flexible connectors on fans at each duct connection. These connectors

should not be pulled taut, but should be long enough to provide folds or flexibility

when the fan is off;

6. Considering floating floors particularly if the mechanical room is located on a

different floor.

The figure below shows a source (air-handling unit) placed on a resilient support inside

the mechanical room.

NOISE FROM FAN SYSTEMS

The most dominating source of noise in HVAC systems is because of the fans and

blowers of the air handling units. The noise generated by fan is transmitted through a

building in different ways: via walls, floors and leakage points to adjacent rooms and via

the supply and extract ductwork to the rooms connected to the ducting.

Types of Fans

The noise generated by a fan depends on the fan type, fan speed, air flow rate, pressure

and the intake/discharge arrangement. HVAC fans are mostly of the centrifugal type or

the axial flow type. The primary selection criterion for a fan is based NOT on its

acoustical characteristics but on its ability to move the required amount of air against the

required pressure. In most cases it is not practical to substitute a fan which generates

less noise, since a quieter design of the same type probably will not meet the other

operating specification for the fan.

Axial-flow fans impart energy to the air by giving it a twisting motion. Axial fans

generate a higher proportion of high frequency noise but less low frequency noise than

centrifugal fans of similar duty. Three types are common:

• Vane-axial fans generally have the lowest amplitudes of sound at low frequency

of any fan. For this reason they are often used in applications where the higher

frequency noise can be managed with attenuation devices. In the useful

operating range, the noise from vane-axial fans is a strong function of the inlet

airflow symmetry and blade tip speed.

• The tube-axial fan generates a somewhat higher noise level than the vane-axial

fan; its spectrum contains a very strong blade frequency component.

• The noise levels of the propeller fan are only slightly higher than those of the

tubeaxial and vaneaxial fans, but such of the noise is at low frequencies and

therefore is difficult to attenuate. Propeller fans are most commonly used on

condensers and for power exhausts.

Centrifugal fans move air by centrifugal action. The centrifugal forces are created by a

rotational air column which is enclosed between the blades of the fans. These produce

most of their noise in the low frequencies, but in general are quieter than axial fans.

Centrifugal fans may be further characterized by the type of blade used. Two common

types of centrifugal fans frequently encountered in HVAC applications are forward

curved and backward curved blades.

• The forward-curved fan is used primarily for HVAC work where high volume flow

rates and low-pressure characteristics are required. Forward curved fans are

commonly used in self-contained package units where space is at a premium.

The most distinguishing acoustical concern of FC fans is the prevalent

occurrence of low-frequency rumble. Forward curved blades will deliver more air

than backward curved blades at a given speed, with increased noise, and if the

resistance is reduced will deliver still more air and may overload the driving

mechanism.

• Backward curved (BC) fans are generally much more energy efficient than FC

fans at higher pressures and airflow. They require higher speed for the same air

delivery and resistance as forward curved blades, but are less noisy and may be

so designed as never to over load when the resistance is reduced. If the fan is

close to a critical space, consider a backward curved (airfoil type) fan and

silencers since the low rumble of forward-curved fans can be difficult to silence.

Many other designs fall between these two options.

Fan Characteristics

Maximum fan efficiency coincides precisely with minimum noise. Select fans that

operate as near as possible to their rated peak efficiency when handling normal airflow

and static pressure. This may seem obvious, but is often overlooked. Using an oversized

or undersized fan can lead to higher equipment noise levels. The resulting variations in

fan RPMs can also lead to airflow fluctuations and duct rumble.

Typical Sound Power Levels of Fans

Sound Power Level, dB (A) at Static Pressure

Volumetric Flowrate (CFM)

0.5 inch wg 3 inch wg

1000 79 95

5000 83 99

10000 85 101

20000 89 105

25000 90 107

50000 93 110

The sound power generated by fan performing a given duty is best obtained from

manufacturer's test data taken under approved (ASHRAE Standard 68- 1986; also

AMCA Standard 300-1986). However, if such data are not readily available, the octave-

band sound power levels for various fans can be estimated by the empirical formulae

described here.

SWL = 77 + 10 log kW + 10 log P

SWL = 25 + 10 log Q + 20 log P

SWL = 130 + 20 log kW – 10 log Q

Where:

• SWL = overall fan sound power level, dB

• kW = rated motor power, kW

• P = static pressure developed by fan, mm w.g.

• Q = volume flow delivered, m3/h

Octave band sound power levels are then found by subtracting correction factors from

the overall sound power level calculated by any one of the above formulae.

Choice of Fan

A low self-noise level is an important criterion when specifying and choosing equipment.

Fans should be chosen so they can operate at high levels of efficiency within their

normal operating ranges. Fans that are made to run at unsuitable operating points, with

subsequent poorer efficiencies, are often noisier than those that have been chosen

correctly.

In CAV (constant flow) systems, fan should be chosen so that their maximum

efficiencies are at the design air flows and static pressure. Operating the fan well off its

duty point will create extra noise. Choosing a high-pressure air mover in a low pressure

application will result in significantly greater noise. Conversely, a low-pressure air mover

heavily loaded in a high-pressure application will also increase noise.

In VAV (variable flow) systems, fans should be chosen so that they can operate with

optimal efficiencies and stability in the most frequently used working ranges. For

example, a fan selected for peak efficiency at full output may aerodynamically stall at an

operating point of 50% of full output resulting in significantly increased low frequency

noise. Similarly, a fan selected to operate at the 50% output point may be very inefficient

at full output, resulting in substantially increased fan noise at all frequencies. The fan for

VAV applications should be selected for peak efficiency at an operating point of around

70 to 80% of the maximum required system capacity. This usually means selecting a fan

that is one size smaller than that required for peak efficiency at 100% of maximum

required system capacity. When the smaller fan is operated at higher capacities, it will

produce up to 5 dB more noise. This occasional increase in sound level is usually more

tolerable than the stall-related sound problems that can occur with a larger fan operating

at less than 100% design capacity most of the time.

A correctly chosen and installed fan reduces the need for noise attenuation in the

ducting system. The following points should be kept in mind:

• Design the systems: the ducts, terminal devices and components for low

pressure drop;

• Compare sound data for different types of fans and from different manufactures

and chose the quietest;

• Choose variable speed control for air flows rather than damper control.

Fan Speed

You need to know two things:

1. Fans are designed to push air: the faster the fan, the more air it pushes

2. Fans produce noise: the faster the fan, the more noise it produces

For our purposes, the best fan is the one that pushes the most air for the least noise.

The effect of rotational speed on noise can best be seen through one of the fan laws:

dB1 = dB2 + 50 log10 (RPM1 / RPM2)

Speed is an obvious major contributor to fan noise. For instance, if the speed of a fan is

reduced by 20%, the dB level will be reduced by 5 dB."

The following table provides a guide to the trade-off that can be expected.

Fan Speed Reduction Noise Reduction

10% 2 dB

20% 5 dB

30% 8 dB

40% 11 dB

50% 15 dB

Fan Selection

While it is true that fan noise is roughly proportional to the 5th power of fan speed, it is

often mistakenly assumed that the lowest RPM fan selection is always the best

selection. Note that the undersized fans operating at higher shaft speeds will produce

more noise, and oversized (larger diameter) fans operating at lower shaft speeds will

create more low-frequency noise (63, 125 and 250 Hz octave bands), which is much

more difficult to remove.

Therefore, when choosing between two different fans of same duty and at same sound

power, the one with the lower sound output at the lower frequencies is preferred from

acoustic point of view. Sound power in higher-frequency bands is easier to control than

low-frequency sound.

Fan discharge velocities for quiet operation

As conditioned air travels from a fan to an occupied room, it is subjected to acceleration,

deceleration, changes in direction, division and a variety of surfaces and obstacles. Each

of these effects disturbs the uniformity of the airflow and causes turbulence, which in

turn creates noise. It is essential to limit the velocity of the airflow through all duct work

systems in order to keep it from generating excessive noise. This is particularly true at

the final branch ducts and the neck of the supply diffusers and return grilles where this

regenerated noise is exposed directly into the occupied spaces.

Air-borne Noise from Fans

The air borne noise from fans mainly comes from the interaction of flow turbulence and

solid surface of fan blades, and blade/fan vibration. The noise is transmitted upstream

and downstream in the connecting ducts or to the atmosphere through the fan case.

Remedies

• Provide flexible connector at the fan discharge;

• Install a silencer at air discharge point of a fan so as to absorb noise generated

from the fan;

• Install acoustic louvers at the fresh air intake;

• Acoustically insulate the mechanical room;

• Divert duct openings away from receivers for outdoor use;

• Provide acoustic enclosure to contain and absorb the noise energy for outdoor

use, if required.

Centrifugal Fan serving the Indoor Spaces

Structure-borne Noise from Fans

Vibration from an operating fan may be transmitted to the interior of the building through

building structure when the fan is directly mounted on a supporting structure without

proper isolation. The vibration noise can be very annoying and must be tackled during

design.

Remedies

• Provide an inertia block to support the fan so as to add rigidity and stability to the

ventilation system;

• Provide vibration isolators to support the inertia block, thereby isolating it from

the building structure;

• Provide flexible connectors between the fan and associated ducts, thereby

isolating it from the ductwork.

Inertia Blocks for Fans

Inertial blocks are often used with fans to reduce the amplitude of oscillation. The table

below shows a recommended weight of inertia block for various rating of ventilation

equipment.

Weight Ratio (Note 1) at Equipment Power Speed

(RPM) Min Area

(Note 2)

Normal Area

(Note 3)

Critical Area

(Note 4)

<3 hp All 2 Centrifugal

Fans 3 – 125 hp <600 2 3

600 -

1200

2

>1200 2

<600 2 3

600 -

1200

2 2

>125 hp

>1200 2 2

Note 1: Weight Ratio – Weight of inertial block over weight of equipment mounted on

inertia block

Note 2: Minimum Area – Basement or on-grade slab location

Note 3: Normal Area – Upper floor location but not above or adjacent sensitive area

Note 4: Critical Area – Upper floor location above or adjacent sensitive areas.

Note: When the mass of the supported equipment is enormous, such as water cooling

towers, there may be no need for additional mass in the form of inertia blocks; a rigid

frame (e.g. Reinforced Concrete Beams) to support the entire assembly may be

sufficient.

Vibration Isolators

Vibration isolators must be matched to the load they carry. A spring that is fully

compressed doesn't offer any isolation and an uncompressed spring is also just as

ineffective. There are many types of off-the-shelf anti-vibration mounts available, for

instance rubber/neoprene or spring types. The type of isolator that is most appropriate

will depend on, among other factors, the mass of the plant and the frequency of vibration

to be isolated. Any supplier of anti-vibration mounts will be able to advise you on this.

Metal Spring:

Springs are particularly applicable where heavy equipment is to be isolated or where the

required static deflections exceed 0.5” (12.5mm). Static deflection of a spring is a value

specified by the suppliers. Selection of appropriate springs is important as this may

result in poor isolation efficiency or even amplification of vibration, especially in the case

that the vibration frequency is extremely low.

The most important feature of spring mountings is to provide good isolation due to its

ability of withstanding relatively large static deflection. Metal springs however have the

disadvantage that at very high frequencies vibration can travel along the spring into the

adhered structure. This is normally overcome by incorporating a neoprene pad in the

spring assembly so that there is no metal-to-metal contact. Most commercially available

springs contain such a pad as a standard feature.

The table below provides the minimum static deflection required for achieving particular

isolation efficiency at different equipment speeds.

Isolation Pads:

Isolation pads can be made of rubber, neoprene, glass fiber or combination of them.

They are relatively cheap, easy for installation and replacement, and have the

advantage of good high-frequencies isolation. However, attention should be given to the

life of the isolation pads as some of them can be damaged by overload or low

temperature.

Guidelines for Centrifugal and Axial Fan Installations

Turbulence results in the generation of noise and an increased static pressure drop in

the system. Therefore, the airflow at the entrance and exit of a fan should be as smooth

as possible to minimize the generation of turbulence. For this reason, fitting (such as

elbows and transitions) should not be placed closer than 3 to 6 duct diameters

downstream from a fan and the duct transitions, particularly at the unit inlet, and

discharge openings should be gradual, 15° maximum to minimize air pressure drop and

turbulence.

The figures below show examples of good and bad airflow conditions for fan

installation:

Guidelines for Centrifugal Fan Installation

Guidelines for Axial Fan Installations

Using Multiple Units

Sometimes, it may be prudent to use several smaller units instead of selecting single

bigger equipment. This reduces the air volume per unit, cutting high air velocities.

Smaller units operating at efficiency point can reduce overall equipment noise, if properly

placed. However, this solution must be carefully planned in advance for several reasons.

First, there must be space for the units and secondly it may be that more equipment

needs to be located closer to the building occupants. Apply discretion while locating the

equipment. Note that the use of in-room or portable air-conditioning units often makes

noise problems worse, not better, for this very reason. Furthermore, multiple units may

cost more initially and result in increased maintenance costs.

Typically a single air handling capacity should be restricted to 30,000 CFM maximum in

commercial buildings. Use of multiple units of smaller capacity is preferred for noise

control as well providing the flexibility of standby operation.

Sound Attenuators for Fans

If lower sound levels are required than are generated by well-designed fans, it is

necessary to add attenuation to the system. This may be provided by sound attenuators

or silencers, which allow the passage of air while restricting the passage of sound

generated from air distribution equipment. They subdivide the airflow into several

passages each lined with perforated sheet backed by mineral wool or glass fiber. A

silencer usually has a cross section greater than the duct in which it is installed such that

noise induced by high air flow velocity passing through the silencer can be avoided.

Silencers are available for circular or rectangular ducts and are fabricated in modular

form in cross section, and in lengths of 0.6, 0.9, 1.2 and 1.5m, etc. They are generally

specified by the insertion loss in decibels (dB) in each octave band, so that the degree of

match with the sound power distribution of the noise source over the frequencies may be

judged. The other important parameter associated with silencers is the resistance to

airflow. The use of silencer will inevitably increase the load of the fan and it is important

to address both the acoustic and air flow performances during the design stage.

The figure below shows a centrifugal fan, with sound attenuators on both the inlet and

the outlet, used in the supply system of a central station ventilating system.

The following facts on silencers are worth noting:

1. Most sound attenuators for fan noise attenuation are of the absorption type

because this type has broad band attenuation characteristics. These silencers

perform well in mid-and high-frequency range and usually consist of sheet steel

duct housing containing sound absorbent ‘splitters’ usually made of rockwool or

glasswool.

2. Silencers restrict airflow, and may require more powerful fans to maintain the

desired air volume. If air flow is restricted too much, the fan works harder, uses

more energy and emits greater sound power levels. However, the less restrictive

the silencer, the less effective it is in reducing low-frequency noise. One means

of countering this problem is to extend the length of the silencer.

3. They are also much larger and heavier than the air ducts, often requiring 6 to 10

feet of unoccupied space. ASHRAE recommends 3-5 duct diameters on either

side of a silencer and the fit-the-system silencers help overcome space

restrictions that most HVAC designers face on a regular basis.

Location of Silencers

The position of the silencer in the duct system can be very important in determining its

effectiveness in reducing the noise at the reception point. The optimum position can be

governed by the possibility of noise breaking into the duct after (downstream as far as

noise is concerned) the silencer, e.g. from noisy plant-rooms, or by possible break-out of

noise from the duct before (upstream of) the silencer. Positioning the silencer close to

bends can cause increases in pressure drop and self-generated noise.

Silencers should ideally be located where the duct leaves the plant room. Care must be

taken to avoid plant room noise from entering the quiet side of the silencer.

If a silencer is located within a mechanical room, noise may enter the system through

the sheet metal duct after the silencer. If a silencer is located away from the mechanical

room walls, noise may escape the system through the sheet metal duct before it is

attenuated by the silencer. Ideally, a sound attenuator should be located within or

immediately adjacent to the mechanical room's walls.

Silencers should be located far enough upstream of any acoustically sensitive space to

ensure that the noise they generate is adequately attenuated before it reaches the

occupied room.

DUCTWORK NOISE

Air flowing through ducts induces vibration at the duct wall, which generates rumbling

noise. In addition, the noise inside the duct can be transmitted to the atmosphere

through the duct surface. To avoid air-borne noise, one should:

1. Size duct for low resistance and for low air velocities. Note that the airflow

generated noise is proportional to the fifth power of the velocity. The general

guidelines are:

Application Supply systems ft/min

Extract systems ft/min

Sound studios, churches, libraries 800-1000 1000-1400

Cinemas, theatres, ballrooms 1000 - 1500 1200-1600

Restaurants, offices, hotels, shops 1200 -1600 1500-1800

In practice, airflow generated noise can be ignored when velocities are below

1500 ft/min in the main duct and 600 ft/min in branch ducts.

2. Smaller ducts tend to be noisier and leakier than larger ducts due to higher air

speeds and pressures. Over-sizing the ducts is a good practice. Figure out the

airflow required for each room in “CFM” (cubic feet per minute). To find out what

size you should make the ducts, divide the CFM by the cross-sectional area of

the duct in square feet (ft²). Example: 500 CFM requiring 12” round duct yields

0.785 ft² cross-sectional area. Therefore, the air flow velocity will be 500/0.785 =

~637 FPM (feet per minute). Any result you get for the above under 1,000 FPM is

good;

3. Avoid unnecessary turbulence by providing adequate distances between

components (at least three duct diameters, but preferably more);

4. Choose components that allow smooth flows through the ducting, bends,

junctions and terminal devices;

5. Avoid sudden cross-sectional changes or sudden changes of flow direction in the

ducting system;

6. Stiffen the vibrating duct surface with supporting webs so as to reduce the

movement of the vibrating surface;

7. Apply damping material to the vibrating duct surface so as to reduce the

movement of vibrating surface;

8. Apply composite lagging of sound absorbing materials to contain the radiation of

noise;

9. With centrifugal-type fans, the low frequency content of the fan noise is the most

prevalent and difficult to attenuate. The best way to attenuates low frequency

noise is by adding mass, i.e. by increasing the duct gauge from SMACNA

standards, one can increase the mass and enhance the sound barrier;

10. The best duct configuration to reduce fan noise breakout is circular duct. The

hoop strength of the circular duct is very good; therefore, minimal noise breaks

out of the ductwork. While we know that most project ceiling plenums cannot

accommodate circular ductwork, the idea in establishing aspect ratios of

ductwork is to size the duct to approach circular;

11. Use round duct near fans when the duct must pass over critical areas. Use

rectangular duct near fans over non-critical areas;

12. Locate main duct systems away from acoustically sensitive areas. The greater

the length of supply duct over non-critical spaces, the more likely the duct is to

reduce its acoustical impact on occupied spaces. Generally, 15 to 20 ft of

ductwork is desired prior to entering the ceiling plenum of the occupied space;

13. Avoid very sharp bends in the ducts. Where bends are necessary, make sure

they are gradual and, if possible, include long, radiuses turning vanes. Use low

pressure drop elbows & fittings (per SMACNA guidelines). Note that the bends

reduce noise, but only if they are gradual and preferably equipped with turning

vanes;

14. If the ducts are sheet metal, you may need to isolate them from the building

using spring isolating hangers;

15. Split larger ducts (>54” wide) into two and if possible, exit the equipment room

using multiple ducts. When the air stream is divided, the sound carried in each

downstream branch is less than the sound upstream of the branch take-off. Thus

if the main duct divides into two equal branches, the sound power level in each

branch just below the junction is 3 dB less than in the main branch just above the

junction. The appropriate level of attenuation can then be determined from table

below.

Duct Splits, dB

% of total air flow

5 10 15 20 30 40 50 60

Attenuation 31 10 8 7 5 4 3 1

If the flow had divided into two parts down one branch and three parts down the

other, then in the smaller branch the branch attenuation would be 10 log10 (5/2) =

4 dB, i.e. the sound power level in the smaller branch would be 4 dB less than in

the main branch. In the other branch the attenuation would be 2.2 dB.

The following figures illustrate some good engineering practices:

Source: SMACNA

Source: SMACNA

Source: SMACNA

Source: SMACNA

Duct Lining

Internal lining of ductwork with suitable sound absorbing materials provides attenuation

to the air-borne noise. The ARI Standard (ARI 885), as well as the ASHRAE Handbook,

provides guidance on insertion loss or decrease in sound power level over a length of

duct.

The table below illustrates why it is important to pay particular attention to lower octave

bands. The insertion loss in the 125 Hz octave band for a 24"X48" duct with 2" lining is 2

dB over a 10’ length of duct. For the same duct, the insertion losses for the 250 Hz and

500 Hz octave bands are 7 dB and 22 dB respectively (see Table below).

Insertion Loss for Common Sized Ducts

Source – ASHRAE 1999, Application Handbook

According to the 1999 ASHRAE HVAC Applications Handbook, this pattern also holds

true for transmission losses through ceiling tile, walls, floors and other building materials.

Consult the Handbook for more detailed information about insertion losses. Generally

speaking, if the equipment sound levels are lower in the first and second octave bands

(63 Hz and 125 Hz); the project will be quiet.

The common duct liner materials are fiber glass, polymide and elastomeric foam. Sound

absorption is directly proportional to exposed surface area and liner thickness. A

common reference for measuring sound absorption of duct liner is the Noise Reduction

Coefficient (NRC). A one-inch-thick fiber glass duct insulation has an NRC of 0.70; that

means it effectively absorbs 70 percent of the sound at the most common frequencies.

Two-inch-thick duct liner has an NRC of 0.90, absorbing 90 percent of the sound.

Note the following facts:

1. The attenuation produced by duct liner (in dB per meter run of duct), depends on

three factors:

a. Sound absorption coefficient (SAC) of the duct lining material: The

attenuation produced by duct lining is greatest at medium frequencies

and poor at low frequencies.

b. Thickness of the absorbing lining: An increase in thickness produce

improved attenuation particularly at low frequencies.

c. Perimeter to cross-sectional area of the duct (P/S): In order to achieve

maximum attenuation the shape of the duct should be designed to give

the highest possible P/S ratio, which in effect means that for a given

cross-sectional area of duct the sound is exposed to the greatest possible

surface area of sound-absorbing material.

2. The location of duct lining can be a critical factor. It is normally placed at the start

of a duct system to attenuate fan noise and near the outlets to correct air flow

generated noise from dampers and fittings.

3. Noise goals depend on the use of a space. An office, apartment, or classroom

usually requires a moderate amount of HVAC-noise reduction for normal

activities to be accommodated. For these projects, as little as 10 to 20 ft of lined

duct, usually immediately downstream from the fan or approximately 10 ft from

fan discharge, is all that is needed for noise goals to be met.

4. Turns and thin linings help reduce high-frequency noise but have little effect on

low rumble. Thick linings, true plenums (large in comparison to connected ducts),

or wraps on rectangular duct provide the most reduction of low-frequency sound.

5. Internal liners reduce duct cross-sectional area. For that reason, liners usually

are kept to a minimum thickness, often 1 inch. To illustrate the impact of liner

thickness, for an 18-in.-by-24-in. duct, going from a 1-in.-thick liner to a 2-in.-thick

liner decreases flow area by about 20 percent, resulting in an increase in air

velocity of 26 percent.

6. For acoustic purposes, square elbows are preferred to radius bends but the

pressure drop will be higher. The lining should have a thickness at least 10% of

D, the clear width between the two linings and the length of lining should extend

a distance not less than 2D before and after the bend.

Unfounded concerns about health issues related to fiber glass being carcinogenic

have been laid to rest by a definitive determination issued by the International

Agency for Research on Cancer, an arm of the World Health Organization.

External Duct Insulation

In addition to the air-borne noise, sound may also transmit directly through the wall of

the ducting and into the surrounding room. This breakout noise can be dampened by

insulating the outside of the duct with a rockwool or glasswool. Further attenuation is

achievable when an additional mass layer is applied as a covering onto the insulation.

The sound attenuation achieved inside the duct is also enhanced by external duct

lagging particularly at low frequencies, up to about 500Hz.

Duct Vibrations

Structure-borne noise often arises due to the mounts (duct hangers or supports)

attaching the ductwork to walls or ceilings and travels through the building in the form of

vibration. When airflow fluctuates, due to a change in fan speed or other variables, it can

make the duct surface vibrate. This rumble by itself can produce sound levels from 65 to

95 decibels at frequencies that range from 10 to 100 Hz. At the mid range, this noise is

as noticeable as normal human speech at a distance of three feet – not easily ignored.

To solve this problem, special flexible fittings should be used which acoustically isolate

the duct system from the structural elements of the building.

Flexible Ducts

Flexible ducts are usually provided to connect the duct to the terminal diffusers, which

are furnished with integral volume dampers. Since dampers generate noise when

partially closed, the sound power levels of the units are a result of the air volume

handled by the diffuser, and the magnitude of the pressure drop across the damper.

A misalignment or offset that exceeds approximately one-quarter diameter in a diffuser

collar length of two diameters can cause a significant change in diffuser sound power

level above that of the manufacturer's published data. The figure below shows an

example of increased pressure drop and increased noise level for a flexible duct

connection. When there is an offset of only 1/8 the diameter, there is no appreciable

change in the diffuser performance.

Duct Fittings

As conditioned air travels from a fan to an occupied room, it is subjected to acceleration,

deceleration, changes in direction, division and a variety of surfaces and obstacles. Each

of these effects disturbs the uniformity of the airflow and causes turbulence, which in

turn creates noise. The phenomenon is more pronounced where rectangular ducts

change size or direction.

1. Elbows, take off fittings, transitions, etc. help reduce low frequency sounds

through the duct system especially in the 500 Hz range as generated by fans.

These duct fittings tend to reflect sounds back upstream, back towards the fan,

thus reducing the amount of sound energy moving forward towards the room.

2. For a square duct, the bend attenuation is a maximum of some 7 or 8 dB at the

frequency (octave band) for which the wavelength of sound in air is twice the

duct width. At higher frequencies, the attenuation drops to 3 or 4 dB and at lower

frequencies it can fall to less than 1 dB.

3. If the bend contains turning vanes to help the airflow around the corner smoothly

then the attenuation produced will be very much reduced (but so, of course, will

the amount of noise regenerated by the bend).

4. Metal turning vanes reduce turbulence by smoothing airflow in the right angle,

but they can create their own noise problems by vibrating and reflecting or

regenerating sound. Choose rounded finger guards when possible.

5. Turns and thin linings help reduce high-frequency noise but have little effect on

low rumble. Thick linings, plenums (large in comparison to connected ducts), or

wraps on rectangular duct provide the most reduction of low-frequency sound.

Duct Crosstalk

Lay ducts to prevent crosstalk and to attenuate the fan noise. The duct work that

connects a fan or air handler to a room is a contained system that will also connect the

equipment noise and vibration to the room unless adequate precautions are taken to

attenuate the noise before it gets there. Without internal sound-absorptive duct liner or

prefabricated sound attenuators, noise travels effectively down the duct system right

along with the conditioned air.

Crosstalk occurs when noise from a space, e.g. talking, music, radiated noise, etc.,

enters the ductwork, propagates along the duct work, and ultimately impacts an adjacent

space. A common example occurs in residential homes when on the third floor, you can

hear the television on the first floor by listening at the supply diffuser or return grille.

Crosstalk is typically composed of ductborne noise, break-in noise, and break-out noise.

The most common method of controlling cross talk is to avoid connecting rooms with

short lengths of duct, by lining the ducts connecting these rooms with acoustical

materials, and by installing silencers (sound absorbing devices) in the duct. Crosstalk

through ducts can also be attenuated by modifications to duct layouts (refer to the figure

below) and increase in the amount of end reflection (higher number of smaller registers

is preferable to fewer larger registers).

Ductwork Layout to Reduce Crosstalk

Although noise control issues through the supply air duct system are routinely

considered in HVAC design, inexperienced mechanical engineers and contractors often

forget that the return-air path is an equally important contributor to noise problems. In

fact, because return-air systems sometimes employ common plenums above corridor

ceilings, there may be less duct work in the return-air path to attenuate the noise, and

the transfer of return air from one space to another may be a significant breach of the

sound isolation between them. The best and really only recommended solution is to

avoid short returns completely. The return intake should be based on a very low face

velocity, return duct lengthened and lined, plus the air should be made to turn at least

twice in its path toward the equipment.

ACOUSTIC LOUVRES

Similar to silencers, acoustic louvers are also commercially available devices that allow

the passage of air while restricting the passage of sound generated from noisy spaces.

They act much the same as ordinary louvers but consist of hollow acoustic vanes

instead of flat sheet vanes or could be fabricated from fiberglass. Turning vanes made of

fiber glass instead of metal eliminate the noise of turbulence and vibration at duct

angles.

The acoustic performance of an acoustic louver is specified by the transmission loss in

decibels (dB) in each octave band. This enables a direct comparison to be made

between the performance of the louver and a solid wall/structure which it probably

replaces. Since an acoustic louver is a very short attenuator, it is appropriate only where

the length of space is restricted and the noise reduction requirement is low. Acoustic

louvers are frequently installed in the facades of buildings where they are architecturally

acceptable and provide a requisite amount of noise attenuation to prevent creating

unacceptably high noise levels outside.

LARGE PLENUMS FOR SOUND ATTENUATION

Plenums can provide a substantial reduction in outlet noise that is critical for achieving

low NC levels outside the MER. An acoustical plenum refers to either an air-handling

unit enclosure or a large volume expansion in the ductwork. The plenum causes little

static pressure drop and is effective in reducing low-frequency noise.

The methods for approximating the acoustic attenuation of a plenum depend on the

sound absorption coefficients of the plenum lining, the exit area, wall surface area, and

the distance between the entrance and exit. It is essential that a plenum with the

following features be part of the overall acoustic solution:

1. 18-gauge outer casing to provide maximum stability and minimize vibration.

2. 3” insulation provides a significantly higher sound absorption coefficient for the

critical lower octave bands versus 2” insulation.

3. 3 lb/ft3 density insulation provides higher sound absorption coefficient for the

critical lower octave bands versus 1½ lb/ft3 density insulation.

4. Perforated sheet metal lining over all insulation to protect it from erosion and

enhance sound attenuation.

5. Factory-supplied duct connections to provide a proper fit and avoid exposed

insulation from field cut duct openings.

6. Internal baffles to minimize turbulence in the plenum, reducing the mechanical

energy and noise.

7. The plenum can be fabricated out of fiberglass board that offers better low-

frequency attenuation than lining. Duct board can “leak” some noise, but does

not carry it along the length of the duct like bare sheet metal. This makes duct

board popular for duct plenums or sections leading directly from the HVAC

equipment. Any noise leaked by the duct board is diffused into unoccupied space

and quiet air is sent on into the system. The use of fiberglass duct is limited by

the levels of static pressure it can withstand and is not advised for high-pressure

systems.

AIR TERMINAL BOXES

VAV (Variable Air Volume) air terminal units are components used in ducted air

distribution systems to maintain the desired temperature in a space served by varying

the volume of air. Basically these units consist of a sheet metal box containing a

damper, controls and a sensor, and they are usually connected to a supply header via a

flexible circular duct. They usually discharge air to one or more diffusers via rectangular

sheet metal ducts. The noise of any air terminal unit can propagate through two

prominent paths:

1. The units discharge ductwork to the connected outlet(s);

2. By direct sound radiated from the casing of the unit into a ceiling plenum and

then through a ceiling (usually acoustical) into the space served.

Usually the VAV installations always require acoustical tile ceiling (NRC times %

coverage > 0.65), but it is of little help if the design does not allow for the source noise

control of VAV boxes. Noise emitted from the opening to the plenum, duct borne supply

and return noise, cabinet radiation, and break-out noise must all be considered along

with attenuation of ductwork.

Quiet VAV Terminal Installation Guideline

Engineers can minimize the sound contribution of air terminals to an occupied space

through good design practice. Typical noise control measures for air terminal units are:

1. Locating units above spaces such as corridors, work rooms, or open plan office

areas. Do not locate VAV units over spaces where the noise should not exceed

an NC or RC 35.

2. The air terminal and the return air grille location should be separated as far as

possible. Radiated sound can travel directly from the terminal through the return

air grille without the benefit of ceiling attenuation. Locating the units at least 5 ft.

away from an open return air grille located in the ceiling.

3. The use of lined duct work or manufacturers’ attenuators downstream of air

terminals can help attenuate higher frequency discharge sound. Flexible duct

(used with moderation) is also an excellent attenuation element. Vinyl flexible

duct is magic for reducing noise. Practically, three or four feet of flex duct can cut

noise as much as 10 or 12 NC points.

4. Designing systems to operate at low supply air static pressure will reduce the

generated sound level. This will also provide more energy efficient operation and

allow the central fan to be downsized.

5. Sharp edges and transitions in the duct design should be minimized to reduce

turbulent airflow and its resulting sound contribution. Installing acoustically lined

sheet metal elbows, at the induced (return) air openings in the casings or

installing acoustically lined elbows above ceiling return air openings when they

must be near, or directly below a terminal unit.

6. Do not place fan-powered boxes of variable volume systems over critical areas.

Locate them over corridors or other non-critical spaces. Make sure there is lining

downstream from such boxes. The same applies to unit heaters and fan coil

units.

7. The air-handling system using VAV boxes must be designed with variable

frequency drive primary fan to take advantage of energy conservation and noise

issues.

8. Minimize inlet static pressure to VAV box; Select VAV box for minimum noise in

125 Hz band (< 65 dB radiated, <70 dB ducted).

9. Large static pressure drops (typically greater than 1 in.) across VAV boxes are

noisy. If possible, design to avoid them. If necessary, provide lined ductwork

and/or silencers downstream.

10. Consult ARI Standard 885, “Procedure For Estimating Occupied Space Sound

Levels In The Application Of Air Terminals And Air Outlets.” This standard

provides current application factors for converting rated sound power to a

predicted room sound pressure level. It also provides a repeatable and

comparable method of both predicting and specifying sound levels.

VAV noise problems are also traced to improper air balancing. For example, air balance

contractors commonly balance an air distribution system by setting all damper positions

without considering the possibility of reducing the fan speed. The end result is a duct

system in which no damper is completely open and the fan is delivering air at a higher

static pressure than would otherwise be necessary. If the duct system is balanced with

at least one balancing damper wide open, the fan speed could be reduced with a

corresponding reduction in fan noise. Lower sound levels will occur if most balancing

dampers are wide open or eliminated.

DIFFUSER, GRILLES AND CONTROL DAMPERS SOUND ISSUES

Diffusers and grilles are devices used to deliver to, or return air from, a building space.

These are the most noise sensitive of all HVAC products since they are almost always

mounted in or directly over occupied spaces. They usually determine the residual

background noise level from 125 Hz to 2,000 Hz.

Generally these devices include vanes, bars, tins, and perforated plates to control the

distribution of air into the space. When the air flows across these elements, unique noise

is generated, that for a particular diffuser or grille design, varies by the 5th to the 6th

power of the velocity. Because of the wide variety in diffuser design, and the sizes

available, manufacturers publish sound level data in their catalogs. Most manufacturers

only provide the NC level that the diffuser noise will reach with different quantities of air

flow in a room. In using the manufacturer’s data care should be taken to note the data

usually applies only to diffusers in an ideal installation. For example placing a damper,

even in an open position, behind a diffuser or grill may increase the noise generated by

up to 15 dB. In addition, to achieve the catalog rating, the duct attached to the diffuser

must be straight for at least 2.5 to 3 equivalent diameters, which is unachievable in many

buildings due to inadequate plenum space.

Good design practices:

1. Use the neck size of the properly selected diffusers as a guide. Flow velocity

within 10 feet of the neck should not exceed the velocity in the neck.

2. Perhaps the most common and potentially greatest source of noise coming from

a diffuser or register is caused by the balancing dampers immediately behind the

diffuser. Depending upon how severely the dampers must be closed to balance

the system, noise levels can increase from 5 to 20 db above ratings.

3. Lining a duct cannot solve a noise problem if the register is the actual source of

the noise, say if fins/bars are vibrating, damper is rattling or air is whistling

through the openings. Generally, if the noise is a rumble, it is the result of a fan

or blower. A whistle on the other hand is a result of high frequency sound.

4. Provide 3 diameters straight flex at diffuser inlet or use minimum 6 feet of

acoustical flex between low pressure duct and diffuser.

5. Select diffuser design and size based on NC rating and number of diffusers:

• 2 diffusers: select NC minus 6

• 4 diffusers: select NC minus 9

• 8 diffusers: select NC minus 12

There is much confusion about diffuser NC ratings. The rating applies to one diffuser in a

room with a "room factor" of usually 10 dB. Beware that some manufacturers use room

factors more than 10 dB in rating their diffusers. Such diffusers are noisier than a diffuser

with the same NC rating based on a 10 dB factor. The room factor is the difference

between the sound power introduced and resulting sound level in the room. It depends

on room geometry and absorption in the room. Most rooms have multiple diffusers and

room factors different from 10 dB. Thus, the catalog rating must be adjusted to get the

expected noise level.

Volume Dampers

An important feature in the air distribution system is its ability to control the amount of air

that flows through each segment of the duct and to ensure that the volume of air

supplied to each space is tailored to its conditioning needs. To accomplish this, volume

dampers are installed to limit the amount of air that is allowed down the duct path.

Unfortunately, dampers accomplish their volume control by pinching down the air

stream, increasing the pressure and consequently the noise wherever they occur.

Ceiling supply diffusers are normally equipped with volume control dampers located right

at the inlet to the diffuser. The airflow noise created by the dampers is essentially

exposed directly into the room and the noise is audible even if the dampers are left wide

open. In acoustically sensitive applications, locate the dampers well upstream of the

outlet within the lined ductwork. The dampers should preferably be located where the

diffuser duct branches off the header, or main duct.

MASKING

In buildings where background sound levels are low, it is possible to alleviate a noise

problem by adding a more pleasant sound to the environment. A good example is open-

plan offices where people can hear activities in other offices and areas. The new sound

has the effect of covering up the noise or making it less noticeable. (An important

caution is that the system be set to operate at not too loud a level or with a harsh

frequency response.)

Many people think of masking as ‘white noise,’ which is characterized by an equal

amount of energy in all audible frequencies. More often, however, masking systems

stress lower frequency sounds, such as the noise produced by a normal HVAC unit. For

example, loudspeakers emitting an HVAC-like sound might be placed between dropped

ceilings and structural ceilings.

In fact, any desirable sound can provide masking. Of nature’s many soothing sounds,

running water is the most popular. Fountains, in fact, have been found to provide both

acoustical and aesthetic enhancements to residential and commercial environments.

CONTROLLING NOISE LEAKS

Penetrations

One of the most commonplace problems caused by HVAC systems has nothing to do

with noise and vibration generated by the equipment or duct work. As it is distributed

throughout the building spaces, duct work inevitably penetrates the wall, ceilings and

floors that are responsible for a room's sound isolation capabilities. If these penetrations

are not adequately treated, they allow sound leaks that can render these acoustical

barriers completely ineffectual. Three treatments are necessary:

• First, the penetrating duct work should be supported independently from the

partition. If a duct rests on the wall it penetrates, any vibration within the duct can

be transmitted into the wall itself, which can then provide a large radiating

surface to turn the vibration into airborne noise.

• Second, the duct should be resiliently isolated from the surrounding construction

as it passes through the partition, to avoid any contact that might transmit the

vibration into the wall.

• Third, the area surrounding the penetration should be sealed airtight, using

materials that won't allow noise from an adjacent space to leak through the gap.

Two other types of noise leaks are common problems in small buildings; leakage

through electrical outlets and leakage around interior partitions can be major.

Electrical and Other Wiring Outlets

Leaks around electrical and other wiring outlets are a common problem, especially when

the outlets are back-to-back. One recommended solution is to offset the boxes. This is

especially effective when the wall is filled with sound absorbing material. Leaks usually

permit only high-frequency sound to pass through. Forcing high-frequency sound to

travel a long path through sound absorbing material is a good way to attenuate it, since

sound absorbing material is most effective at high frequencies.

The use of blocking panels is another way of achieving good sound insulation while still

having back-to-back outlets.

Partitions and Ceiling Spaces

This major leak is common with suspended ceilings. Sound is transmitted via the space

above the ceiling where the common wall does not extend to the slab above. There are

two approaches to dealing with this problem: either the partition is made full height or the

attenuation of the path through the plenum and ceiling is increased. For each of these

approaches there are variants.

Noise leak through plenum space

Sound leaks can be controlled using sound barrier materials, ceiling boards and other

sound absorbing materials over the suspended ceiling.

Partition extension is another way but remember that extending the wall will interfere

with airflow and ventilation when the space above the ceiling is used as a return air

plenum. Additional ductwork penetrating the barrier will be necessary to restore the

airflow. Since this ductwork introduces a path for sound, it should be lined with sound

absorbing material.

USE OF ACTIVE NOISE CANCELLATION EQUIPMENT

As one of the noise control methods discussed so far, active noise control involves

electronically altering the character of the sound wave in order to reduce its level. In

sound cancellation, a microphone measures the noise and a processor generates a

mirror image (180° out of phase) of its source. This mirror image is then broadcast in the

path of the original sound. Depending on the circumstance, the new sound can cancel

enough of the original signal to reduce noise levels up to 40 decibels.

Although sound cancellation is a very powerful noise control tool, it is only practical in

confined environments where tonal frequencies are below 500 Hertz (cycles per

second). Ventilation ducts are ideal candidates for active noise control because they are

enclosed, with the dominant noise often consisting of low-frequency pure tones

(associated with the fan characteristics).

Noise cancellation equipment requires 10 to 20 feet of space and can easily cost 10

times what a silencer costs. Nonetheless, in the right situation, it is less expensive than

tearing out walls, ceilings, and plumbing to install soundproofing. The system is

commonly used in efforts to silence return ducts. The system can be installed in tight

places, offers superior silencing of low-frequency noise and causes no static pressure

drop.

CONTROLLING NOISE FROM OUTDOOR EQUIPMENT

Air-borne Noise from Air-cooled Chillers

Noise generated from air-cooled chillers may cause noise disturbance to nearby

residents. It mainly comes from the air flow noise resulting from air turbulence at

condenser fans and compressor noise during running and on/off cycle of refrigerant.

Typical Sound Power Levels of Air-cooled Chillers

Cooling Capacity (Ton) Sound Power Level, dB (A)

50 100

100 102

150 103

200 105

250 106

300 106

350 107

The other problem is the structure-borne noise. The vibrations due to chiller operation

can be transmitted indoors through building structure at points where the chiller is rigidly

fixed to the structure without proper isolation. The vibration transmitted may activate the

building structure to generate noise which can be annoying to the occupants of the

building.

Remedies

• Use quieter chiller models. If the total sound power level of the machine exceeds

100 dB(A), erect a barrier or partial enclosure between the plant and nearby

surroundings to block the noise propagation path;

• Apply barrier or partial enclosure all around when there are noise sensitive

receivers in the surroundings;

• Install floating floor so as to reduce air-borne noise transmission through floor

slab when the floor underneath is a noise sensitive receiver;

• Provide vibration isolators to support an air-cooled chiller.

Air-borne Noise from Water Cooling Towers

Noise from water cooling towers mainly comes from the air flow noise resulting from air

turbulence at condenser fans, mechanical noise of fan(s) and motor(s) and water

splashing noise due to water flowing through the tower into the collection basin.

Typical Sound Power Levels of Water Cooling Towers

Horsepower of Fan (hp) Sound Power Level, dB (A)

10 96

20 99

30 101

40 102

50 103

60 104

80 105

Remedies

• Erect a barrier or partial enclosure between the plant and nearby surroundings to

block the noise propagation path;

• Provide acoustic mat on the water surface so as to reduce the water splashing

noise;

• Install acoustically lined vent cowl at fan discharge outlet;

• Apply partial enclosure all around when there are noise sensitive receivers all

around;

• Provide vibration isolators to support water cooling tower, thereby isolating it from

the building structure.

REDUCING NOISE TRANSMISSION FROM OUTDOOR MACHINES

Sound enclosures, barriers and partitions reduce sound transmission between adjacent

spaces.

Complete Enclosures

When a noise reduction of 20dB (A) or more is required, it is generally necessary to use

a complete enclosure especially if the noise problem is a result of air-borne noise

transmission. The enclosure should be internally lined with 2 inch thick sound absorbing

material (e.g. fiber glass).

Ventilation of enclosures should not be overlooked as most equipment, such as motors,

air cooled condensers or cooling towers require an adequate air supply either to prevent

overheating or to enable them to function efficiently.

Partial Enclosures

Partial enclosures are structures erected around a source of noise, but not fully

enclosing the source and leaving space for natural ventilation, which will be effective

only when there is no line of sight between the noise source and the receiver. The use of

partial enclosures has advantages over complete enclosures in terms of cost,

accessibility, and ventilation, but design and construction should be done carefully.

Ideally, a reduction of up to 20dB (A) can be achieved.

Barriers

To be effective, an acoustic barrier needs to be placed as close as possible either to the

noise source or the receiving position. There should be no gap or joint in the barrier

through which noise will leak. The surface density of the barrier must be at least

10kg/m2. Ideally, the length of the barrier should be at least 5 times its height. Line of

sight between the source and the receiver must be cut off completely.

A reduction of noise level of between 5dB (A) to 10dB (A) can generally be resulted.

Noise reduction will be greater if the barrier is lined with sound absorbing material at the

surface of the barrier facing the noise source or is extended as high as possible above

the line of sight.

ADDITIONAL THOUGHTS

Beyond the initial design, mechanical engineers can try many things to reduce unwanted

noise from HVAC systems. Thus the foremost thing is to downsize HVAC equipment and

reduce airflow, if possible.

Correctly Estimating the Heat Loads

When designing an HVAC system, the first step is to accurately identify the heat loads

generated by the occupants, equipment, lighting and surrounding environment. Although

this isn't really an acoustical issue, it's an area that suffers from the "garbage in, garbage

out" syndrome. If a project's mechanical and electrical engineers are given bad

information about equipment loads or the heat dissipation, the HVAC systems will have

a high mismatch between capacity and actual load.

Most of the time, rules of thumb data is used to size HVAC equipment. For instance,

heat load is often made on sq-ft basis without considering the geographical location and

volume of the space. Engineers often base HVAC sizing decisions on the full nameplate

or “connected” load of computers, copiers, printers, and so on; and assume

simultaneous operation of such equipment. In fact, most of this equipment operates at a

fraction of the nameplate value, and rarely operates simultaneously. Many HVAC

designs are based on plug load assumptions on the order of 5 W/sq-ft in office spaces.

Invariably, these do not reflect the true operating environment and in actual practice bear

little resemblance to their actual power consumption over time. According to an

ASHRAE, one W/sq-ft is a reasonable upper bound when equipment diversity and

reasonable estimates of the true running load are included.

Over and above, for conservation reasons, the HVAC designers keep very high level of

safety margin.

Many factors affect heating or cooling load. A good estimator will measure walls,

ceilings, floor space, and windows for the accurate determination of room volume and

also accounts for the true R-value of the building insulation, windows, and building

materials. A close estimate of the building's air leakage is necessary and a good

estimate will also include an inspection of the size, condition (how well joints are sealed

and the ducts are insulated), and location of the distribution ducts. Make sure the

designer uses the correct design outdoor temperature and humidity for your area.

The following may be noted:

Insist upon a correct system sizing calculation before signing a contract. This service is

often offered at little or no cost to homeowners by gas and electric utilities, major heating

equipment manufacturers, and conscientious heating and air conditioning contractors.

Manual J, published by the Air Conditioning Contractors of America (ACCA), is the most

common method in use. Many user-friendly computer software packages or worksheets

can simplify the calculation procedure.

Using the correct inputs shall result in right load estimation and optimum size of air

handling & refrigeration equipment.

Reducing Airflow

The greater the airflow the greater is the noise. A 20% reduction in airflow will reduce the

noise level by approximately 5 dB. Understanding some of the fan fundamentals

provides clues to where to look for possible reduction of air flow and consequent

reduction in equipment sizes. Fan airflow in CFM is a function of the heat load and also

the temperature differential of the supply and return air temperature. For a given air-

conditioning load, as the supply air temperature is reduced, the supply air volume is

reduced proportionally. The sensible heat gain equation is Q = 1.08 × CFM × ∆T

Where:

• Q is sensible heat in Btu/hr

• CFM is the air volume required

• ∆T is the temperature differential of the space setpoint minus the supply air

temperature

Let’s check this for 1 ton of air-conditioning load. (Note that 1 ton of refrigeration, Q is

equivalent to heat extraction rate of 12000 Btus per hour.)

Consider two cases:

Case # 1: The room setpoint temperature is 75°F and the supply air temperature is 55°F

Case # 2: The room setpoint temperature is 75°F and the supply air temperature is 45°F

In case # 1, the ∆T is 20°F and therefore the air volume per ton of air-conditioning load

shall be:

CFM = 12000 / (1.08 x 20) or = 555/ton

In case # 2, the ∆T is 30°F and therefore the air volume per ton of air-conditioning load

shall be:

CFM = 12000 / (1.085 x 30) or = 370/ton

This shows that by lowering the supply-air temperature from the 55°F to 45°F reduces

the supply-air volume by 33%. The air handling units, fan terminal units, etc., are

reduced to approximately one half the size which means the fan/s and motor/s shall be

smaller and therefore it lowers the mechanical sound power levels of the air handling

units. Thus the cold air systems deliver much less air volume and are substantially

quieter on air sound, often minimizing, or entirely eliminating, sound attenuating

requirements.

______________________________________________________________________

Summarizing…..

In HVAC system design, controlling three types of sound propagation are important:

1. Air-borne noise is transmitted through both interior partitions and the exterior

facade. Sources include mechanical rooms (interior), rooftop equipment, and

fans exhausting to the exterior via louvers or stacks. Mechanical equipment noise

can also come from neighboring buildings. Control of air-borne mechanical noise

can be achieved using appropriate partition construction and detailing, building

facade design, site planning, silencers, acoustic louvers, barriers, and selection

of quieter equipment.

2. Duct-borne noise travels efficiently through ventilation ducts to any space that is

serviced by the system. Duct-borne noise is best reduced at the source by

selecting quieter fans or by adding silencers. Duct layouts should incorporate

space for silencers. Duct layouts should enhance privacy and sound isolation.

3. Structure-borne noise is caused by vibration from mechanical equipment entering

the structure where it can propagate efficiently and be re-radiated as air-borne

noise. Structure borne noise is best controlled at the source using rubber or

spring isolators and inertia bases.

Majority of the noise in HVAC systems is attributed to the air distribution and the fan

systems. The noise control strategy for HVAC systems can be described by the following

eight steps.

1. Select a suitable location for mechanical equipment rooms away from noise-

sensitive areas of the building.

2. Select fans with low sound power output levels. Ideally the octave band sound

power levels should be known for both the sound radiated into the duct and that

from the fan casing into the fan room.

3. Vibrationally isolate the fan. This includes both isolating the fan and motor from

the floor, and installing a vibration break in the duct immediately adjacent to the

fan, to prevent the propagation of vibration energy through the duct walls.

4. Acoustically isolate the fan noise. The equipment room walls, floor, and ceiling

must provide a high transmission loss to the airborne noise in the equipment

room. One must attenuate the sound propagating down the duct by the

installation of a dissipative silencer immediately after the duct vibration break.

Such a silencer is selected to provide adequate attenuation in each octave band

and also to not produce excessive airflow noise.

5. Calculate the attenuation of the sound propagating through the duct system so

that the sound power entering each room is known.

6. Estimate flow noise sound power levels entering each room. Sharp bends,

control mechanisms, grills, diffusers, and combinations of flow-noise-producing

devices placed too close together will lead to increased flow-induced noise. Flow-

induced noise is very dependent on the velocity of the air flow and can thus be

dramatically reduced by reducing this velocity.

7. Calculate the combined sound power entering the room and the expected octave

band sound pressure levels in each room. From these, calculate an NC value

and compare it with the established noise criteria for the room.

8. If the desired noise criteria are exceeded, further reductions are needed. If fan

noise propagating through the duct is the dominant source of noise in the room,

an improved dissipative silencer would help. If flow-induced noise is the dominant

problem, lining the last few meters of the duct with sound-absorbing material

would help. This would require a slightly larger duct to maintain the original

internal dimensions and flow velocities. Reducing the air flow velocity or

modifying particular flow-noise-producing devices could also be considered.

If the above guidelines are followed properly, it will reduce the noise level significantly.

The target noise level from a HVAC system must be chosen with care after studying the

scope and the area of application. Remember there is no justification in reducing noise

levels below reasonable values. In fact, the total cost of a system increases

exponentially as the NC rating is lowered. For this reason, it is recommended to consult

an acoustic engineer for critical applications.

______________________________________________________________________

Annexure - 1

RULES OF THUMB

1. When specifying sound criteria for HVAC equipment “refer to sound power level,

not sound pressure level.”

2. When comparing sound power levels, remember the lowest and highest octave

bands are only accurate to about +/-4 dB.

3. Lower frequencies are the most difficult to attenuate.

4. 2 x sound pressure (single source) = +3 dB (sound pressure level)

5. 2 x distance from sound source = -6dB (sound pressure level)

6. +10 dB (sound pressure level) = 2 x original loudness perception

7. When trying to calculate the additive effect of two sound sources, use the

approximation (note that logarithms cannot be added directly).

If two sound levels are identical, the combined sound is three dB higher than either. If

the difference is 10 dB, the highest sound level completely dominates and there is no

contribution by the lower sound level.

General rules of thumb for controlling noise within a space:

1. You have to at least double the absorption in a space before there is a noticeable

difference. Every time you double the absorption, the reverberant noise field is

reduced by 3 dB, which is classified as “just perceptible”.

2. Adding absorption to a space can prove a clearly noticeable improvement, if the

space is fairly reverberant to begin with. The practical limit for noise reduction

from absorption is 10 dB, which sounds half as loud.

3. The improvement will not be as noticeable as you get closer to the noise source.

4. Carpet is not a cure. In fact, it is typically only 15-20% absorptive. It would take

four times as much carpet to have the same impact as a typical acoustic

material, which is about 80% absorptive.

General rules of thumb for controlling noise between spaces:

1. A wall must extend to the structural deck in order to achieve optimal isolation.

Walls extending only to a dropped ceiling will result in inadequate isolation.

2. Sound will travel through the weakest structural elements, which, many times,

are doors, windows or electrical outlets.

3. When the mass of a barrier is doubled, the isolation quality (or STC rating)

increases by approximately 5 dB, which is clearly noticeable.

4. Installing insulation within a wall or floor/ceiling cavity will improve the STC rating

by about 4-6 dB, which is clearly noticeable.

5. Often times, specialty insulations do not perform any better than standard batt

insulation.

6. Metal studs perform better than wood studs. Staggering the studs or using dual

studs can provide a substantial increase in isolation.

7. Increasing air space in a wall or window assembly will improve isolation.

Facts on dB

1. Human range of hearing starts at 0 dB and is considered safe up to 70 dB. Over

and above this level, it is hazardous and can result in permanent hearing

damage.

2. Auditory nerves can be permanently damaged from prolonged exposure at 90

dB.

3. 120 dB can cause pain and ringing in the ear.

4. A change in level of 10 dB corresponds approximately to a doubling of perceived

loudness or a 5-decibel reduction can cut the risk of hearing loss in half.

5. On the decibel scale, a soft whisper at 2 m distance would have a sound level of

about 35 dB (A) while average background sound levels in an office would be

about 40 dB (A).

6. A jack hammer 15 m away can raise the sound level to 95 dB (A), while a

discotheque can generate noise levels in excess of 110 dB (A).

7. The human ear is not equally sensitive at all frequencies; sounds of the same

level but with different frequencies will not be considered equally loud. For

example, a sound at 3 kHz and a level of 54 dB will sound about as loud as one

at 50 Hz and a level of 79 dB.

8. Doubling the sound power raises the sound power level by 3 dB.

9. Doubling the sound pressure raises the sound pressure level by 6 dB.

10. Depending on the frequency there is a difference on how noise is perceived. For

example, for low frequencies (10-100 Hz) the upper pain limit is 130-140 db, for

100-1000 Hz the limit is 120-125 db, and for high frequencies (1000-10000 Hz)

the limit is 110-130 db.

11. The sound intensity is attenuated in relation to the distance from the source. For

example, it is reduced by 10 db at a distance of 3 m and by 40 db at a distance of

100 m from the source.

12. A basic measure of sound, the sound pressure level (SPL) is expressed in

decibels (dB). Sound pressure is the parameter that is normally measured in

noise assessments. People's hearing mechanisms respond to pressures that

represent the range from the threshold of hearing to the threshold of pain. When

the SPL = 0 dB, the acoustic pressure is the same as the threshold of hearing.

13. OSHA Noise Standard indicates that for sound levels of:

• 84 dB A or less - No hearing protection required

• 85-89 dB A - Hearing protection highly recommended

• 90 dB A or greater - Hearing protection is required at the workplace

14. 20 dB is not twice as loud as 10 dB. Because a logarithmic scale is used, every

3 dB increase in sound level doubles the perceived sound level for humans. If the

energy level is 20, the dB measure is 13 dB.

15. Not all sound pressures are equally loud. The two identical sound pressures but

at different frequencies shall be perceived to be at a different loudness. This is

because the human hearing process is not the same at all frequencies.

______________________________________________________________________

Annexure - 2

GLOSSARY OF NOISE TERMS

Absorption Coefficient: A measure of the quantity of sound lost on impinging on a

surface. It can be defined as: {1 – (Reflected sound energy / Incident sound energy)}

It is a property of the material on which the sound impinges and is dependant on

thickness of material and frequency of the sound.

Absorption: The properties of a material composition to convert sound energy into heat

thereby reducing the amount of energy that can be reflected.

Absorptive Silencer: A silencer making use of the absorptive properties of materials to

reduce the sound passing through it.

Acoustics: The science of sound; its production, transmission and effects.

Acoustical: The properties of a material to absorb or reflect sound.

Acoustical Analysis: A review of a space to determine the level of reverberation or

reflected sound in the space (in seconds) as influenced by the building materials used to

construct the space. Also, a study of the amount of acoustical absorption required to

reduce reverberation and noise.

Acoustical Environment: The acoustical characteristics of a space or room influenced

by the amount of acoustical absorption, or lack of it, in the space.

Ambient Noise: The existing background noise in an area can be sounds from many

sources, near and far.

Anechoic Room: A specially constructed room in which as much sound as possible is

absorbed at its boundaries. It is typically achieved by using sound absorbing wedges.

Architectural Acoustics: The control of noise in a building space to adequately support

the communications function within the space and its effect on the occupants. The

qualities of the building materials used determine its character with respect to distinct

hearing.

Articulation Class (AC): A single number rating used for comparing acoustical ceilings

and acoustical screens for speech privacy purposes. AC values increase with

increasing privacy and range from approximately 100-250. This classification

supersedes speech privacy Noise Isolation Class (NIC) rating method.

Articulation Index (AI): A measure of speech intelligibility influenced by acoustical

environment rated from 0.01 to 1.00. The higher the number the higher the intelligibility

of words and sentences understood from 0-100%.

Area Effect: This term suggests the efficiency of sound absorption. Acoustical materials

spaced apart can have greater absorption than the same amount of material butted

together. The increase in efficiency is due to absorption by soft exposed edges and also

to diffraction of sound energy around panel perimeters.

Attenuation: The reduction of sound energy as a function of distance traveled.

Attenuator: It is a noise-reducing device - often colloquially known as a 'silencer'.

A-Weighing: An electronic filtering system in a sound meter that allows meter to largely

ignore lower frequency sounds in a similar fashion to the way our ears do.

Ambient Noise/Sound: Noise level in a space from all sources such as HVAC or

extraneous sounds from outside the space. Masking sound or low-level background

music can contribute to ambient level of sound or noise.

Audiogram: Graph of hearing threshold level as a function of frequency (ANSI S3.20-

1995: audiogram).

Audiometer: An instrument for measuring hearing acuity.

Barrier: A material that when placed around a source of noise inhibits the transmission

of that noise. Also, anything physical or an environment that interferes with

communication or listening e.g., a poor acoustical environment can be a barrier to good

listening and especially so for persons with a hearing impairment.

Background Noise: The existing noise associated with a given environment, can be

sounds from many sources, near and far. (See also Ambient Noise.)

Baffle: A free hanging acoustical sound absorbing unit. Normally it is suspended

vertically in a variety of patterns to introduce absorption into a space to reduce

reverberation and noise levels.

Baseline Audiogram: The audiogram obtained from an audiometric examination

administered before employment or within the first 30 days of employment that is

preceded by a period of at least 12 hr of quiet. The baseline audiogram is the

audiogram against which subsequent audiograms will be compared for the calculation of

significant threshold shift.

BEL: A measurement of sound intensity named in honor of Alexander Graham Bell. First

used to relate intensity to a level corresponding to hearing sensation

Boominess: Low frequency reflections. In small rooms, acoustical panels with air space

behind can better help control low frequency reflectivity.

Breakout: It is the escape of sound from any source enclosing structure, such as

ductwork and metal casings.

Cloud: In acoustical industry terms, an acoustical panel suspended in a horizontal

position from ceiling/roof structure; similar to baffle but in a horizontal position.

Cocktail Party Effect: Sound in a noisy crowded room generated mostly by

conversation. Levels rise and fall as people compete with one another to be heard.

Perception of speech can be nearly impossible in high levels of noise.

Cochlea: A snail shaped mechanism in the inner ear that contains hair cells of basilar

membrane that vibrate to aid in frequency recognition.

Continuous Noise: Noise with negligible small fluctuations of level within the period of

observation (ANSI S3.20-1995: stationary noise; steady noise).

Crest Factor: Ten times the logarithm to the base ten of the square of the wideband

peak amplitude of a signal to the time-mean-square amplitude over a stated time period.

Unit dB (ANSI S3.20-1995: crest factor)

Cross talk: It is the transfer of airborne noise from one area to another via secondary air

paths, such as ventilation ductwork or ceiling voids.

Cut off Frequency: It is the frequency at which performance of an acoustic item or

material starts to fall below normal or below criterion. Applied to anechoic wedge

treatment it refers to the frequency below which the absorption coefficient is worse than

0.99.

Cycle: In acoustics, the cycle is the complete oscillation of pressure above arid below

the atmospheric static pressure.

Cycles per second: The number of oscillations that occur in the time frame of one

second. Low frequency sounds have fewer and longer oscillations.

Decibel (dB): One tenth of a Bel; a Bel being a unit of amplification corresponding to a

tenfold increase. In terms of Sound Level Measurements it is related to datum levels as

follows:

• For Sound Pressure Level (SPL) datum = 2 X 10-5 Pascal

• For Sound Power Level (SWL) datum = 1 X 10-12 Watts

• S.P.L. (dB) = 20 log [P / (1 x 10 -5)], where P is the amplitude of the Sound

Pressure Waves concerned measured in Pascal.

• S.W.L. (dB) = 10 log [W / (1 x 10-12)], where W is the sound power radiated by

the source concerned measured in watts.

Decibel, A-Weighted (dBA): Unit representing the sound level measured with the A-

weighting network on a sound level meter.

Decibel, C-Weighted (dBC): Unit representing the sound level measured with the C-

weighting network on a sound level meter.

Deaf: Loss of auditory sensation with or without use of assistive listening device

Diffusion: The scattering or random reflection of a sound wave from a surface. The

direction of reflected sound is changed so that listeners may have sensation of sound

coming from all directions at equal levels.

Directivity Factor: When sound radiates from any source sound levels can be higher in

certain directions than others. This is called 'Directivity'. Directivity Factor is the ratio of

the increased level to the average value.

Directivity Index: Is directivity factor expressed in decibels (dB).

It is usually designated by DIO where O is the angle between the axis of the source and

the direction of the measuring point.

Discrete Frequency: It is a single frequency signal or a single frequency noise

sufficiently dominant over other frequencies to be distinctly audible.

Dose: The amount of actual exposure relative to the amount of allowable exposure, and

for which 100% and above represents exposures that are hazardous. The noise dose is

calculated according to the following formula:

D = {C1/T1 + C2/T2 +...+ Cn/Tn} H 100 where Cn = total time of exposure at a specified

noise level Tn = exposure time at which noise for this level becomes hazardous

Dynamic Insertion Loss (DIL): It is a measure of the acoustic performance of an

attenuator when handling the rated flow. Not necessarily the same as Static Insertion

Loss because it may include regeneration and / or other velocity effects and will account

for the effects of the actual fluid and fluid conditions for which the silencer is designed.

Echo: Reflected sound producing a distinct repetition of the original sound. Echo in

mountains is distinct by reason of distance of travel after original signal has ceased.

Echo Flutter: Short echoes in small reverberate spaces that produce a clicking, ringing

or hissing sound after the original sound signal has ceased. Flutter echoes may be

present in long narrow spaces with parallel walls.

Effective Noise Level: The estimated A-weighted noise level at the ear when wearing

hearing protectors. Effective noise level is computed by (1) subtracting de-rated NRR

from C-weighted noise exposure levels, or (2) subtracting de-rated NRR minus 7 dB

from A-weighted noise exposure levels. Unit, dB

End Reflection: End reflection occurs when sound energy radiates from a hole. The

sudden expansion to atmosphere causes some low frequency noise to be reflected back

towards the source. Expressed in decibels (dB), the effect is dependent on hole size and

frequency. Maximum at lowest frequency from smallest hole

Equal-Energy Hypothesis: A hypothesis stating that equal amounts of sound energy

will produce equal amounts of hearing impairment, regardless of how the sound energy

is distributed in time.

Equal Loudness Contours: Curves represented in graph form as a function of sound

level and frequency which listeners perceive as being equally loud. High frequency

sounds above 2000 Hz are more annoying. Human hearing is less sensitive to low

frequency sound. (See also Phon).

Equivalent Continuous Sound Level: Ten times the logarithm to the base ten of the

ratio of time-mean-square instantaneous A-weighted sound pressure, during a stated

time interval T, to the square of the standard reference sound pressure. Unit, dB

Exchange Rate: An increment of decibels that requires the halving of exposure time or

a decrement of decibels that requires the doubling of exposure time. For example, a 3-

dB exchange rate requires that noise exposure time be halved for each 3-dB increase in

noise level; likewise, a 5-dB exchange rate requires that exposure time be halved for

each 5-dB increase.

Free Field: It is a sound field, which is free from all reflective surfaces or where there

are no obstructions. A simulated free field can be produced inside an anechoic room.

Frequency: For a function periodic in time, the reciprocal of the period. Unit, hertz (HZ)

(ANSI S1.1-1994: frequency)

Frequency (Hz) – Sound: It is the number of sound waves to pass a point in one

second.

Frequency – vibration: It is the number of complete vibrations in one second.

Frequency Analysis: An analysis of sound to determine the character of the sound by

determining the amount of sounds of various frequencies that make up the overall sound

spectrum i.e. higher frequency sound or pitch vs. low frequency.

Flanking transmission: It is the transfer of sound between any two areas by any

indirect path, usually structural. It can also apply to noise transmitted along the casing of

a silencer.

Hearing Impairment: Hearing impairment means a hearing loss of mild, moderate or

severe degree as opposed to "deafness" which is generally described as little or no

residual hearing with or without the aid of an assistive listening device

Hearing Range:

• 16 - 20000 Hz (Speech Intelligibility)

• 600 - 4800 Hz (Speech Privacy)

Hearing Threshold Level (HTL): For a specified signal, amount in decibels by which

the hearing threshold for a listener, for one or both ears, exceeds a specified reference

equivalent threshold level. Unit, dB (ANSI S1.1-1994: hearing level; hearing threshold

level)

Helmholz Resonance: A resonance created by the mass of a "plug" of fluid acting on

the resilience of "spring" of a volume of fluid; e.g. a "plug" of air in a bottleneck resonates

on the volume when one blows across the neck. This principle can be used in silencers

etc.

Hertz (Hz): Frequency of sound expressed by cycles per second.

Impulse: Product of a force and the time during which the force is applied; more

specifically, impulse is the time integral of force from an initial time to a final time, the

force being time-dependent and equal to zero before the initial time and after the final

time (ANSI S1.1-1994: impulse).

Impulsive Noise: Impulsive noise is characterized by a sharp rise and rapid decay in

sound levels and is less than 1 sec in duration. For the purposes of this document, it

refers to impact or impulse noise.

Insertion Loss: The reduction of noise level by the introduction of noise control device

established by the substitution method of test, or by "before and after" testing. The term

can be applied to all forms of treatment including silencers and enclosures. (See also

Dynamic Insertion Loss and Static Insertion Loss)

Insulation (Sound): It is the property of a material or partition to oppose sound transfer

through its thickness.

Inverse Square Law: The reduction of noise with distance in terms of decibels, it means

a decrease of 6dB for each doubling of distance from a point source when no reflective

surfaces are apparent. This is only applied in free field conditions where the source is

small in comparison with the distance.

Intensity: (See loudness).

Intermittent Noise: Noise levels that are interrupted by intervals of relatively low sound

levels.

Inverse Square Law: Sound levels full off with distance traveled. Sound level drops off

6 dB from source point for every doubling of distance.

Loudness: A listener's auditory impression of the strength of a sound. The average

deviation above and below the static value due to sound wave is called sound pressure.

The energy expended during the sound wave vibration is called intensity and is

measured in intensity units. Loudness is the physical resonance to sound pressure and

intensity.

Laminar Flow: It is colloquially used to describe the preferred state of airflow. Strictly

means undisturbed flow at very low flow-rates where the air moves in parallel paths.

Level Difference: The difference in Sound Pressure Levels between two positions, e.g.

inside and outside an enclosure. (This is not the same as Insertion Loss, Transmission

Loss or Sound Reduction Index although in some circumstances they may be similar.)

Masking: The process by which extra sound is introduced into an area to reduce the

variability of fluctuating noise levels or the intelligibility of speech.

Mass Law: Heavy materials stop more noise passing through them than light materials.

For any airtight material there will be an increase in its "noise stopping" ability of

approximately 6dB for every doubling of mass per unit area.

Near Field: It is the area close to a large noise source where the inverse square law

does not apply.

Noise: Undesired or unwanted sound. By extension, noise is any unwarranted

disturbance within a useful frequency band, such as undesired electric waves in a

transmission channel or device.

Noise Criterion Curves (NC): An American set of curves based on the sensitivity of the

human ear. They give a single figure for broadband noise. It is used for indoor design

criteria.

Noise Rating Curves (NR): A set of curves based on the sensitivity of the human ear.

They are used to give a single figure rating for a broad band of frequencies. It is used in

Europe for interior and exterior design criteria levels. They have a greater decibel range

than NC curves.

Noise Isolation Class (NIC): A single number rating of the degree of speech privacy

achieved through the use of an Acoustical Ceiling and sound absorbing screens in an

open office. NIC has been replaced by the Articulation Class (AC) rating method.

Noise Reduction: The amount of noise that is reduced through the introduction of

sound absorbing materials. This is established by measuring the difference in sound

pressure levels adjacent to each surface

Noise Reduction Coefficient (NRC): The NRC of an acoustical material is the

arithmetic average to the nearest multiple of 0.05 of its absorption coefficients at 4 one-

third octave bands with center frequencies of 250, 500, 1000, 2000 Hertz.

Noise Reduction Rating (NRR): The NRR, which indicates a hearing protector’s noise

reduction capability, is a single-number rating that is required by law to be shown on the

label of each hearing protector sold in the United States. Unit, dB

Octave: A pitch interval of 2:1. The tone whose frequency is twice that of the given tone

Octave Bands: It is a convenient division of the frequency scale. Sounds that contain

energy over a wide range of frequencies are divided into sections called bands.

Identified by their center frequency, typically 63 125 250 500 1000 2000 4000 8000 Hz.

PHON (Loudness contours): A subjective impression of equal loudness by listeners as a

function of frequency and sound level (dB). An increase in low frequency sound will be

perceived as being much louder than an equivalent high frequency increase.

Pink Noise: It is noise of a statistically random nature, having an equal energy per

octave bandwidth throughout the audible range.

Pitch: The perceived auditory sensation of sounds expressed in terms of high or low

frequency stimulus of the sound.

Presbycusis: The loss of hearing due primarily to the aging process. High frequency

loss is frequently a result of early hearing loss.

Pressure Drop: The difference between the pressure upstream and down stream of a

silencer at given flow conditions. If the silencer is to be installed other than in a duct

system of constant cross-section care must be taken with regard to measuring positions

and methods to allow for difference in velocity head.

Pulse Range: Difference in decibels between the peak level of an impulsive signal and

the root-mean-square level of a continuous noise.

Pure Tone: It is a single frequency signal.

Random Noise: A confused noise comprised from large number of sound waves, all

with unrelated frequencies and magnitudes.

Reactive Attenuator or Resonant Attenuator / Silencer: An attenuator, in which the

noise reduction is brought about typically by changes in cross section, chambers and

baffles sections.

Reflection: The amount of sound wave energy (sound) that is reflected off a surface.

Hard non porous surfaces reflect more sound than soft porous surfaces. Some sound

reflection can enhance quality of signal of speech and music.

Regeneration: The noise generated by airflow turbulence. The noise level usually

increases with flow speed.

Resonance: The emphasis of sound of a particular frequency.

Resonant Frequency (Hz): It is the frequency at which resonance occurs in the resilient

system.

Reverberation: Reflected sound in a room, which decays after the sound source has

stopped. Reverberation Room or Chamber: A calibrated room specially constructed

with sound reflective walls, e.g., plastered concrete. The result is a room with a "long

smooth echo", in which a sound takes a long time to die away. The sound pressure

levels in this room are very even.

Reverberation Time: The time taken for sound to decay 60 dB to 1 / 1,000,000 of its

original sound level after the sound source has stopped. Sound after it has ended will

continue to reflect off surfaces until the wave loses enough energy by absorption to

eventually die out. Reverberation time is the basic acoustical property of a room, which

depends only on its dimensions and the absorptive properties of its surfaces and

contents. Reverberation has an important impact on speech intelligibility.

Room Constant: It is the sound absorbing capacity of a room, usually expressed in m2.

Sabin: It is a unit of absorption comprising the sum of the products of absorption

coefficients and areas of the materials of a room. It must be qualified by the units of area

used e.g. Sq-m Sabines.

Sabine Formula: A formula developed by Wallace Clement Sabine that allows

designers to plan reverberation time in a room in advance of construction and

occupancy. Defined and improved empirically, the Sabine Formula is T = 0.049(V/A)

where T = Reverberation time or time required (for sound to decay 60 dB after source

has slopped) in seconds. V = Volume of room in cubic feet. A = Total square footage of

absorption in sabins. It becomes inaccurate when absorption is high.

Septum: A thin layer of material between 2 layers of absorptive material i.e. Foil, lead,

steel, etc. that prevents sound wave from passing through absorptive material.

Signal to Noise Ratio: The sound level at the listener’s ear of a speaker above the

background noise level. The inverse square low impacts on the S/N ratio. Signal to

Noise Ratios are important in classrooms and should be in range of + 15 to +20 dB.

Significant Threshold Shift: A shift in hearing threshold, outside the range of

audiometric testing variability (5 dB), that warrants follow-up action to prevent further

hearing loss. NIOSH defines significant threshold shift as an increase in the HTL of 15

dB or more at any frequency (500, 1000, 2000, 3000, 4000, or 6000 Hz) in either ear

that is confirmed for the same ear and frequency by a second test within 30 days of the

first test.

Silencer: It is colloquialism for attenuator.

Sound: Oscillation in pressure, stress, particle displacement, particle velocity, etc. in a

medium with internal forces (e.g., elastic or viscous), or the superposition of such

propagated oscillations.

Sound Absorption: The property possessed by materials, objects and air to convert

sound energy into heat. Sound waves reflected by a surface cause a loss of energy. The

energy not reflected is called its absorption coefficient.

Sound Absorption Coefficient: The fraction of energy striking a material or object that

is not reflected. For instance, if a material reflects 70% of the sound energy incident

upon its surface, then its Sound Absorption Coefficient would be 0.30. SAC =

absorption/area - sabins per sq. ft.

Sound Insulation: It is the property of a material or partition to oppose sound transfer

through its thickness.

Sound Intensity: Average rate of sound energy transmitted in a specified direction at a

point through a unit area normal to this direction at the point considered. Unit, watt per

square meter (W/m2); symbol, I (ANSI S1.1-1994: sound intensity; sound-energy flux

density; sound-power density)

Sound Intensity Level: Ten times the logarithm to the base ten of the ratio of the

intensity of a given sound in a stated direction to the reference sound intensity of 1

picoWatt per square meter (pW/m2). Unit, dB

Sound Level: A subjective measure of sound expressed in decibels as a comparison

corresponding to familiar sounds experienced in a variety of situations.

Sound Level Meter (Noise Meter): It is an instrument for measuring sound pressure

levels. It can be fitted with electrically weighting networks for direct read-off in dBA, dBB,

dBC and octave or third octave bands.

Sound Power: It is a measure of sound energy in watts. It is a fixed property of a

machine, irrespective of environment.

Sound Power Level (SWL or PWL): It is the amount of sound output from a machine,

etc., cannot be measured directly. It is expressed in decibels of SWL.

Sound Pressure: Root-mean-square instantaneous sound pressure at a point during a

given time interval. Unit, Pascal (Pa); (ANSI S1.1-1994: sound pressure; effective sound

pressure)

Sound Pressure Level (SPL): It is a measurable sound level that depends upon

environment. It is a measure of the sound pressure at a point in N/m2. It is expressed in

decibels of SPL at a specified distance and position. It can also be considered as a

measure of intensity of terms of Sound Energy per unit area at the point considered, but

is not a vector (i.e. directional) as Sound Intensity strictly is.

SPL Direct Field: It is the sound radiating directly from the source(s) to the receiver

without reflection.

SPL Reverberant Field: It is the sound reaching the receiver after one or more

reflections.

Sound Level Meter: A device that converts sound pressure variations in air into

corresponding electronic signals. The signals are filtered to exclude signals outside

frequencies desired.

Sound Reduction Index (SRI): A set of values measured by a specific test method to

establish the actual amount of sound that will be stopped by the material, partition or

panel, when located between two reverberation rooms. Average SRI can be calculated

by averaging the set of values in the sixteen third octave bands from 100 Hz to 3150 Hz.

It is a property of the material(s) or construction, not directly measurable in the field.

Sound Transmission Class (STC): The preferred single class rating system designed

to give the sound insulation properties of a structure for the rank ordering of a series of

structures

Spectrum: The description of a sound wave's components of frequency and amplitude.

Sound Spectrum: It is the separation of sound into its frequency components across

the audible range of the human ear.

Speech: The act of speaking. A child learns to speak by imitating those people around

him. It is important that a child can hear proper speech. 'We speak what we hear.'

Speech Intelligibility: The ability of a listener to hear and correctly interpret verbal

messages. In a classroom with high ceilings and hard parallel surfaces such as glass

and tile, speech intelligibility is a particular problem. Sound bounces off walls, ceilings

and floor, distorting the teacher's instructions and interfering with students' ability to

comprehend.

Speech Privacy: The degree to which speech is unintelligible between offices. Three

ratings are used: Confidential, Normal (Non Obtrusive), Minimal.

Standing Waves: These occur due to room geometry. Sound levels at some locations

in the room at certain frequencies will be intensified by additive interference of

successive waves, and in other locations reduced by cancellation.

Static Insertion Loss (SIL): The Insertion Loss of an attenuator under static (no flow)

conditions (c.f. Insertion Loss, Dynamic Insertion Loss).

Threshold Shift: A partial loss of hearing caused by excessive noise, either temporary

or permanent, in a person's threshold of audibility.

Threshold of Audibility or Hearing: The minimum sound levels at each frequency that

a person can just hear.

Threshold of pain: The sound level at which a person experiences physical pain.

(Typically 120 dB)

Third Octave Bands: It is a small division of the frequency scale, three to each octave.

It enables more accurate noise analysis.

Time Weighted Average (TWA): The averaging of different exposure levels during an

exposure period. For noise, given an 85-dBA exposure limit and a 3-dB exchange rate,

the TWA is calculated according to the following formula: TWA = 10.0 H Log(D/100) + 85

where D = dose.

Tinnitus: 'Ringing in the ears' of which there is no observable cause.

Transmission Loss: American preferred description for sound reduction index. A set of

values measured by a specific test method to establish the actual amount of noise that

will be stopped by the material, partition or panel when placed between two

reverberation rooms.

Turbulent Flow: A confused state of airflow that may cause noise to be generated

inside, for example, a ductwork system.

Varying Noise: Noise, with or without audible tones, for which the level varies

substantially during the period of observation

Velocity Head or Velocity Pressure (Pv): It is a measure of the inertia of a flowing fluid

used in assessing pressure losses in duct systems and/or silencers for air at

atmospheric conditions.

Pv = v2/4; where, Pv is in millimeters of water and v is the velocity in meters/seconds.

Or

Pv = v2 / 3970; where, Pv is in inches of water and v is the velocity in feet/minute.

Volume: The cubic space of a room bounded by walls, floors, and ceilings determined

by Volume = length x Width x Height of space. Volume influences reverberation lime.

Wavelength: Sound that passes through air produces a wavelike motion of compression

and rarefaction. Wavelength is the distance between two like points on a wave shape,

e.g. distance from crest to crest. Length of sound wave varies with frequency. Low

frequency equals longer wavelengths.

White Noise: It is noise of a statistically random nature having an equal energy level per

Hertz throughout the audible range.

*****

Annexure – 3

Sound Power & Sound Power Levels

The table below indicates the Sound Power and the Sound Power Level from some

common (and some not so common) sources.

Sound Power (Watts)

Sound Power Level (dB)

(re 10-12 W)

Source

25 to 40,000,000 195 Saturn Rocket

100,000 170 Jet engine

10,000 160 Turbojet engine 3200kg thrust

1000 150 4 propeller airliner

100 140 Large centrifugal fan, 800000 m3/h

10 130 • 75 piece orchestra

• Axial fan, 100000 m3/h

1 120 • Large chipping hammer

• Human pain limit

0.1 110

• Large aircraft 150 m over

head

• Centrifugal fan, 25000 m3/h

• Blaring radio

Sound Power (Watts)

Sound Power Level (dB)

(re 10-12 W)

Source

0.01 100

• Large air compressor

• Air chisel

• Magnetic drill press

• High pressure gas leak

• Banging of steel plate

• Drive gear

• Car on highway

• Normal fan

0.001 90

• Axial ventilating fan (2500

m³/h)

• Cut-off saw

• Hammer mill

• Small air compressor

• Grinder

• Heavy diesel vehicle

• Heavy city traffic

• Lawn mover

• Maximum sound up to 8 hour

• OSHA1) criteria - engineering

or administrative noise

controls)

Sound Power (Watts)

Sound Power Level (dB)

(re 10-12 W)

Source

• Jackhammer at 15 m

• Bulldozer at 15 m

0.0001 80

• Voice - shouting

• Maximum sound up to 8 hour

(OSHA criteria - hearing

conservation program)

• Pneumatic tools at 15 m

• Alarm clock

• Buses, trucks, motorcycles at

15 m

• Dishwasher

0.00001 70

• Voice - conversational level

• Car at 15 m

• Vacuum cleaner at 3 m

0.000001 60 • Large department store

• Busy restaurant or canteen

0.0000001 50 Room with window air conditioner

0.00000001 40 Voice, low

0.000000001 30 • Voice - very soft whisper

Sound Power (Watts)

Sound Power Level (dB)

(re 10-12 W)

Source

• Room in a quiet dwelling at

midnight

0.0000000001 20 Noise at ear from rustling leaves

0.00000000001 10 Quietest audible sound for persons

under normal conditions

0.000000000001 0

Quietest audible sound for persons

with excellent hearing under

laboratory conditions

______________________________________________________________________